Objective: Based on magnetic resonance imaging of patients with brain metastases, using imaging omics methods to differentiate the primary lesion of brain metastases and determine the specific source of the patient’s brain metastases. Methods: The data of this study were collected from patients with brain metastasis who treated in Liaoning Cancer Hospital from January 2017 to September 2020. Experienced imaging doctors manually delineate the active area of the patient’s brain metastasis tumor, and obtain a 4 mm circular area containing the active area and the surrounding area through computer methods. And 1 967 imaging omics features were extracted from each region of the contrast-enhanced T1-weighted imaging (CE-T1WI) and T2-weighted imaging (T2WI) sequences. After merging the sequences, a three-step feature screening method was performed using U-test, least absolute shrinkage and selection operator (LASSO) logistic regression, and Akaike information criterion (AIC), draw the receive operating characteristic (ROC) curve to calculate the area under curve (AUC) value as an indicator of the classification performance of the discriminative model. Results: A total of 215 patients with brain metastases were identified, including 100 cases of lung origin, 50 cases of breast origin, 50 cases of gastrointestinal origin, and 15 cases of other origin. From each of the two sequences, 1 967 imaging histological features were extracted. Three radiomics models were successfully constructed to differentiate whether brain metastases arose from lung cancer lesions in the RS_Lung model, breast cancer lesions in the RS_breast model and gastrointestinal cancer lesions in the RS_Gastrointestic model, with training set AUC of 0.898, 0.872 and 0.938, sensitivity of 0.908, 0.744 and 0.860, specificity of 0.818, 0.879 and 0.909, respectively. Conclusion: This study achieved good results in the distinguishing task derived from brain metastases using an imaging omics model constructed based on a 4 mm annular region, which has the potential to be a noninvasive preoperative new marker and guide personalized treatment options for patients with brain metastases.

Prostate specific membrane antigen (PSMA) was first discovered in prostate cancer cells and is an important target for the diagnosis and treatment of tumors with high expression of PSMA such as prostate cancer. The development and application of a variety of imaging and therapeutic radionuclide-labeled targeted PSMA ligands play an important role in the theranostics of prostate cancer. This article reviewed the relevant research and clinical applications theranostics of PSMA.

The application of fibroblast activation protein (FAP) in cancer-associated fibroblast (CAF) has garnered widespread attention in nuclear medicine. FAP is highly expressed in most cancer tissues and rarely expressed in healthy tissues. Therefore, radionuclide probes targeting FAP have significant diagnostic and therapeutic potential. Compared to  ¹⁸ F-FDG, widely used in clinical practice, FAP-targeted tracers exhibit superior target-to-background ratio (tumor-to-background ratio) across various indications. Unlike  ¹⁸ F-FDG, the clinical use of FAP-targeted tracers does not require complex preparations like dietary restrictions for patients and offers the possibility of radioligand therapy (RLT). While radiolabeled antibodies have been studied clinically since the 1990s, the introduction and clinical application of FAPI tracers in 2018 marked a breakthrough event in the development of FAP-targeted probes within nuclear medicine. Since then, the development and application of targeted FAP tracers have become a hot topic among pharmaceutical companies as well as within the radiopharmaceutical and nuclear medicine communities. This article reviewed the current status and research progress of FAP-targeted radiopharmaceuticals.

 In recent years, China has made significant progress in the field of liver ultrasound research, providing more accurate and reliable means for clinical diagnosis and treatment. Reviewing the clinical research on liver ultrasound in China in 2023, the hotspots in clinical research on diffuse liver diseases are mainly the quantitative assessment of fatty liver and the diagnosis of liver fibrosis with new ultrasound technologies. The clinical research on focal liver diseases focuses on the precise diagnosis of liver tumors and the related applications of artificial intelligence technology. In the field of diagnosis and treatment of liver disease with minimally invasive interventional ultrasound, it covers the optimization of ablation treatment strategies in liver tumors, the risk assessment of post-ablation complications, and prognosis evaluation, etc. The achievements in the field of liver guidelines and consensus in 2023 will also be introduced in this article.

The molecular imaging technology of nuclear medicine is widely used in the diagnosis of tumors, neurological diseases, cardiovascular diseases and other diseases due to its non-invasiveness and high sensitivity. Fibroblast Activation Protein (FAP) is a type Ⅱ serine protease that is highly expressed in 90% of malignant solid tumors. A variety of FAP inhibitors (FAPI) have been developed to target FAP, most of which have nanomolar levels of FAP affinity as well as high FAP selectivity for positron emission tomography(PET)/computed tomography(CT) imaging of tumors. In addition to this, FAP is also highly expressed in activated fibroblasts, and some scholars have explored the value of FAPI PET/CT in non-tumor diseases. In this paper, we reviewed the progress of FAP-targeted molecular imaging probes and research in oncologic and non-oncologic diseases.

Objective: To explore the clinical and imaging features of diffuse high-grade gliomas in the brainstem of children based on multi parameter sequences, so as to improve the comprehensive understanding of the disease. Methods: The clinical and imaging data of pediatric brain stem diffuse high-grade gliomas confirmed by pathology from January 2022 to January 2024 were retrospectively analyzed. Results: A total of 45 children in Xinhua Hospital, Shanghai Jiao Tong University School of Medicine with diffuse high-grade gliomas in the brainstem (19 males and 26 females, aged 6.48±2.34 years) were included in this study. Among them, 4.4% (2/45) had cerebrospinal fluid dissemination, while the rest had localized or nearby infiltration and growth in the brainstem. 45 cases underwent computed tomography (CT) plain scan+ magnetic resonance imaging (MRI) plain scan/enhanced scan+diffusion-weighted imaging (DWI) basic scan, 6 cases underwent CT perfusion imaging (CTP) scan simultaneously, 8 cases underwent magnetic resonance spectroscopy (MRS) scan simultaneously, 7 cases underwent diffusion tensor imaging (DTI) scan simultaneously, and 2 cases underwent susceptibility weighted imaging (SWI) scan simultaneously. All 45 cases had endogenous swelling growth, including 36 cases with uniform swelling growth on both sides and 9 cases with lateral swelling growth (7 cases crossed the midline during the growth process, and 2 cases with diffuse growth on one side of the head and tail); 28.9% (13/45) limited brainstem growth, 2.2% (1/45) limited medullary growth, 68.9% (31/45) grew towards the head and tail (midbrain/thalamus/medulla oblongata, 21/31, 67.7%) and laterally (pons/cerebellum, 10/31, 32.3%), and 26.7% (12/45) progressed to exogeneity (including 11 cases involving the basilar artery, 11/12, 91.7%). The CT plain scan sequence showed low-density thickening shadows in the brainstem area, with a density lower than that of the normal brainstem (P<0.001). CT perfusion sequence scan showed a decrease in cerebral blood flow (CBF) in the tumor area of 83.3% (5/6) of patients (1 case showed a significantly lower than average level in the tumor center area), and 1 case showed an increase. Bilateral cerebral hemisphere perfusion abnormalities, with delayed peak time, mainly around the lateral ventricles. The MRI structural sequence showed that 8.9% (4/45) of the tumor solid components on T1-weighted imaging (T1WI) showed equal signal intensity, 11.1% (5/45) showed low signal intensity, and 80.0% showed mixed signal intensity (36/45). The T2-fluid suppressed inversion recovery (FLAIR) sequence showed slightly high to high signal intensity, and the pattern after T1WI enhancement showed diverse changes. Six cases did not show enhancement, while the rest showed varying degrees of mild to moderate enhancement. 28.2% (11/39) showed local circular enhancement, 38.5% (15/39) showed local nodular enhancement, 10.3% (4/39) showed patchy enhancement, and 23.1% (9/39) showed patchy enhancement. Manifested as gravel like enhancement, 7.7% (3/39) showed mixed mode enhancement, and some patients with unclear T1WI enhancement had better T2-FLAIR enhancement sequence contrast. In the MRI functional sequence, MRS showed varying degrees of increase in Cho peak and decrease in NAA peak, with one case showing a significant inversion of Lac peak; DTI scan showed that the nerve fiber bundles in the lesion area were pushed to the periphery by the tumor, with 5 cases showing partial interruption of the fiber bundles and 1 case only showing compression and pushing of the fiber bundles. Local low signal can be seen on the SWI phase map, indicating intratumoral microbleeds. Conclusion: The conventional CT findings of diffuse high-grade gliomas in the brain stem of children are low-density brain stem thickening, and quantification is helpful for early detection and prevention of missed diagnosis. The growth mainly occurred in the diffuse endogenetic pontine, and most of them infiltrated and grew to other brain regions along the direction of nerve fiber bundles, pushing or destroying the nearby fiber bundles. The enhancement patterns were various, and most of them were scattered mild to moderate enhancement. There may be cystic image signals inside, which may be related to the reduction of necrosis by local perfusion and absorption after microbleeds;

This review summarized the progress of clinical research on liver disease ultrasound in China in 2024, covering multiple fields, including precise diagnosis of focal liver lesions, prediction of biological characteristics of hepatocellular carcinoma (HCC), evaluation of therapeutic efficacy and prognosis, quantitative diagnosis of metabolic-associated fatty liver disease (MAFLD), and interventional ultrasound treatment. In the field of liver tumor diagnosis, the application scope of the contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) has been further expanded, with significant improvement in diagnostic performance. The integration of artificial intelligence and multimodal data has played an important role in predicting the biological characteristics of HCC, enhancing the stability and generalizability of predictions. Ultrasound technology combined with multimodal models has shown significant advantages in evaluating the efficacy and prognosis of liver tumors. Additionally, ultrasound fat quantification techniques have achieved important breakthroughs in the diagnosis of MAFLD, transitioning from single-center to multi-center validation. In interventional ultrasound research, the application of irreversible electroporation and artificial intelligence-assisted navigation systems has significantly improved the precision and safety of interventional treatments. Overall, the research on liver ultrasound in China in 2024 has actively explored the application of artificial intelligence and multimodal data integration, significantly improving the accuracy and efficiency of liver disease diagnosis and treatment evaluation. Significant progress has been made in the directions of multi-center validation and precision medicine, providing new ideas and strategies for clinical practice.

Objective: Radioligand therapy (RLT) is an emerging modality that combines targeted therapy with radiation-induced cytotoxicity. This study investigates the efficacy and safety of  ¹⁷⁷ Lu-fibroblast activation protein (FAP)-2286 RLT in patients with advanced lung cancer. A comprehensive assessment was conducted through clinical trials and patient outcome analyses to evaluate the therapeutic potential of this treatment, aiming to improve the prognosis of patients with advanced lung cancer. Methods: The study included patients with advanced lung cancer treated with  ¹⁷⁷ Lu-FAP-2286 RLT in the Affiliated Hospital of Southwest Medical University from September 2022 to April 2024. Baseline scans using  ⁶⁸ Ga-FAP-2286 positron emission tomography (PET)/ computed tomography (CT) confirmed high expression of FAP within the lesions. Each treatment cycle administered approximately 200 mCi (7.4 GBq) of  ¹⁷⁷ Lu-FAP-2286 via intravenous infusion over 4 h, with intervals of 8-12 weeks between cycles, for a total of 4-6 cycles. The evaluation of safety and efficacy included laboratory tests, adverse event documentation, and assessments according to response evaluation criteria in solid tumor (RECIST) 1.1 and adopted PET response criteria in solid tumor (PERCIST) 1.0 criteria. Data analysis was performed using SPSS version 26.0, with paired sample t-tests applied to compare pre- and post-treatment differences following normality tests. Results: Among the 10 patients included in the study (8 males and 2 females, aged 53-73), treatment cycles varied: one patient received six cycles, three patients received four cycles, one patient received three cycles, four patients received two cycles, and one patient received only one cycle. The median follow-up period was 13 months. Treatment with  ¹⁷⁷ Lu- FAP-2286 was well-tolerated, with the primary adverse effects being fatigue and abdominal distension. No severe hematological toxicity or significant impairment of liver and kidney function was observed. Efficacy evaluation revealed a partial response (PR) rate of 30.0%, a stable disease (SD) rate of 50.0%, and a progressive disease (PD) rate of 20.0%. The overall response rate was 30.0%, and the disease control rate was 80.0%. According to the adopted PERCIST 1.0 criteria, the PR rate was 30.0%, the SD rate was 50.0%, and the PD rate was 20.0%. Conclusion:  ¹⁷⁷ Lu-FAP-2286 RLT demonstrated good safety and significant efficacy in patients with advanced lung cancer, with high disease control and overall response rates. Although the sample size in this study was small, the results support  ¹⁷⁷ Lu-FAP-2286 RLT as a promising new strategy for treating advanced lung cancer. Further research involving larger sample sizes and longer follow-up periods is required to validate its efficacy and safety.

Objective: To investigate the value of multiple radiomics models constructed by combining clinical risk factors and multiparametric magnetic resonance imaging (MRI) radiomics features to predict microsatellite instability (MSI) in rectal cancer. Methods: A total of 149 rectal cancer patients were included in the First Affiliated Hospital of Bengbu Medical University from December 2020 to November 2023, including 34 patients with MSI and 115 patients with microsatellite stability (MSS). Based on MRI examination of multiple sequence images, 3D regions of interest were delineated, radiomics features were extracted and dimensionality was reduced to select the best features. Then, five different machine learning algorithms, including logistic regression (LR), random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), and naive Bayes tree (NBT) were used to construct different imaging radiomics models using the optimal radiomics features. And receiver operating characteristic (ROC) curves were drew to evaluate the diagnostic performance of different models. Results: The RF model showed the most stable performance, and the clinical imaging radiomics joint model nomogram based on clinical independent risk factors and imaging radiomics showed high diagnostic efficiency for MSI in rectal cancer. The area under curve (AUC) of the training group and the validation group were 0.923 and 0.914, respectively, indicating the most significant evaluation of MSI in rectal cancer. Conclusion: Combining different machine learning algorithms, a clinical imaging radiomics nomogram constructed from clinical risk factors and multi-parameter MRI radiomics features can effectively predict the unstable state of preoperative rectal cancer.

The Chinese Anti-Cancer Association Society of Integrative Cardio-oncology, the Ultrasound Branch of the Chinese Medical Association, and the Chinese Society of Echocardiography convened experts formulated the Chinese guideline for the clinical application of noninvasive imaging technology in accessing cancer therapy-related cardiovascular toxicity (2023 edition). The guideline highlighted that cancer therapy-related cardiovascular toxicity (CTR-CVT) has become one of the major threats to the survival of cancer patients. Non-invasive imaging technologies play crucial roles in the risk stratification, early diagnosis, monitoring and follow-up of CTR-CVT. This article provided an interpretation of the non-invasive imaging examination methods and monitoring strategies for common CTR-CVT as proposed in the guideline, offering practical and effective guidance for the clinical management of CTR-CVT.

 In the year 2023, high-frequency ultrasound technology has seen significant progress in the field of skin diseases. It has made notable achievements in both basic research and clinical applications. Particularly, it has been successful in skin disease diagnosis, pre- and post-treatment evaluations, artificial intelligence-based diagnosis, the transformation of basic research results, and new tumor treatment methods. This technology provides dermatologists with more accurate and multi-dimensional clinical diagnosis and treatment information. This article provided a comprehensive review of the latest progress in skin ultrasound technology in 2023.

Objective: To investigate the feasibility of fusion registration of high-frequency ultrasound (HFUS)-pathology images and predictive value of model for predicting the pathological tissue components in in vitro models based on deep learning (DL) networks. Methods: Sixty in vitro mimetic tumor models containing four different biological tissues and localized particles were prepared. HFUS images and wide slide images (WSI) of the same slide were obtained under physical registration. The obtained images were quality controlled and selected, and the region of interest (ROI) was manually outlined along the edges of the tissues in the WSI and then transferred to HFUS images. The datasets were consisted of original images and corresponding ROIs and were divided into the training set (n=462), validation set (n=34) and testing set (n=38) at the ratio of 13∶1∶1. DL models were developed via transfer learning DeepLabV3, FCN-50 and MobileNetV3 networks. The pixel accuracy (PA), precision, recall and F1-score were used to quantify and compare the performance of each model in the training and testing datasets. The automatically segmented images were output. Results: The DL models based on DeepLabV3, FCN-50 and MobileNetV3 networks had high accuracy and similarity for automatically segmentation of different tissue components in the testing set, and the MobileNetV3 model outperformed others with the PA of 91.4% and F1-score of 87.1%. There was no significant difference between performance of models (all P> 0.05). There were statistically significant differences between the efficiencies of models for predicting different in vitro biological tissue components (all P<0.001), with the best of liver tissue. Conclusion: The constructed ultrasound-pathology fusion models in this study can effectively recognize the in vitro tissue components in ultrasound images and provide the methodological basis for further clinical applications.

Objective: To evaluate and compare the diagnostic value of contrast-enhanced ultrasound (CEUS), ultrasound-guided fine-needle aspiration biopsy (US-FNAB), and the combination of the two in the diagnosis of Thyroid Imaging Reporting and Data System (TI-RADS) 4 nodules of the thyroid gland. Methods: Patients with TI-RADS 4 thyroid nodules having clear pathological findings from February 2019 to November 2023 in Tangshan Central Hospital and Tangshan Gongren Hospital were retrospectively analyzed. The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of CEUS, US-FNAB, and the combination of the two methods were calculated by using postoperative pathology as a criterion, and the diagnostic efficacy of CEUS, US-FNAB and their combination was compared by drawing the receiver operating characteristic (ROC) curve. Results: In the differential diagnosis of seventy-five patients benign and malignant thyroid TI-RADS 4 nodules, the diagnostic sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of CEUS were 70.7%, 70.6%, 70.6%, 74.4%, and 66.7%, respectively, and of US-FNAB were 75.6%, 79.4%, 77.3%, 81.6% and 73.0%, respectively, and the combination of the two were 97.6%, 58.8%, 80.0%, 74.1%, and 95.2%, respectively. The false-positive rates of CEUS, US-FNAB, and the combination of the two were 29.3%, 24.4%, and 2.4%, respectively. The false-negative rates of CEUS, US-FNAB and the combination of the two were 29.4%, 20.6% and 41.2%, respectively. The AUC of CEUS, US-FNAB and the combination of the two applied diagnostics were 0.707, 0.775, and 0.782, respectively, as derived from plotting the ROC curve. Conclusion: Both CEUS and US-FNAB have a certain value in the differential diagnosis of benign and malignant thyroid TI-RADS 4 nodules, and their diagnostic value can be improved when they are combined, which can be combined when it is difficult to distinguish between benign and malignant in the clinical situation.

In the contemporary medical field, the incidence of cancer is on the rise, and cancer therapy - related cardiovascular toxicity (CTR-CVT) has become a crucial issue that cannot be overlooked, having a significant impact on the prognosis of patients. With the increasing diversification and sophistication of cancer treatment methods, including chemotherapy, radiotherapy, targeted therapy, and immunotherapy, while these offer hope for anti-tumor treatment to patients, they also carry the potential risk of cardiovascular toxicity, affecting treatment effectiveness and patients’ quality of life. Therefore, it is of particular importance to accurately and comprehensively assess CTR-CVT. This article evaluated cardiovascular toxicity from multiple dimensions such as clinical manifestations, electrocardiogram, imaging, and biomarkers, aiming to provide precise and practical assessment strategies for clinical practice. These strategies are intended to balance the efficacy of cancer treatment and cardiac safety, improve patients’ quality of life, optimize long-term outcomes, and enhance patients’ living standards.

Objective: To explore the clinical application value of the contrast-enhanced ultrasound (CEUS) scoring method in reclassifying Breast Imaging Reporting and Data System (BI-RADS) category 4 lesions. Methods: The data of patients with breast nodules diagnosed by postoperative pathology or puncture biopsy in Shanghai Yangpu District Shidong Hospital and Shanghai Fourth People’s Hospital from May 2021 to April 2024 were retrospectively analyzed. All nodules underwent routine ultrasound examination and were indicated as BI-RADS category 4. Each nodule underwent CEUS and was assigned a score. The BI-RADS classification was adjusted according to the score results, and the diagnostic efficiency before and after adjustment was compared. In addition, breast nodules were divided into 3 groups according to the maximum diameter, and the diagnostic efficacy of CEUS scoring was evaluated for nodules of different sizes. Results: A total of 105 patients with 109 nodules were included, comprising 59 benign nodules and 50 malignant nodules. BI-RADS 4b and above were considered as malignant, and the sensitivity and accuracy of diagnosis combined with CEUS scoring method (94.00% and 92.66%) were significantly higher than those of conventional ultrasound (52.00% and 75.23%, P<0.05). The area under the receiver operating characteristic (ROC) curve after adjustment was 0.95, significantly higher than that of conventional ultrasound (0.76). After integrating the CEUS scoring method, the diagnostic sensitivity significantly improved across all three groups of breast nodules stratified by size (P<0.05). The diagnostic accuracy in all three groups increased to over 90%, with the most notable improvement observed in the ≥2 cm group (65.38% vs 96.15%, P< 0.05). Conclusion: The CEUS scoring method has high diagnostic efficiency and is valuable for the readjustment of BI-RADS category 4 nodules, and has extensive clinical application value.

Objective: To evaluate the relationship between International Association for the Study of Lung Cancer (IASLC) classification system and traditional computed tomography (CT) imaging features, and to construct a hierarchical prognosis model based on CT imaging features. Methods: The study retrospectively analyzed the medical records of 102 consecutive patients with primary pathological (p) stage I (T1N0M0 or T2aN0M0) LUAD in Nanjing Chest Hospital from January 2019 to May 2022. According to the 2020 IASLC grading system, patients were classified and the clinical pathological and imaging features were compared between different IASLC histological grades, as well as between recurrent and non recurrent groups. Logistic regression analysis was used to determine CT signs related to IASLC grading, and to determine influencing factors for disease-free survival (DFS) of patients through a multivariate Cox regression model. Results: A total of 102 patients with LUAD were divided into grade 1 (15 cases, 14.7%), grade 2 (63 cases, 61.8%) and grade 3 (24 cases, 23.5%). During the 30.4 months follow-up, 16 patients (15.7%) relapsed. Higher CTR (OR=2.152, 95% CI 1.530-3.264, P=0.005) and higher CT value (OR=3.730, 95% CI 2.841-6.353, P=0.001) were independent risks of higher histological grade. The AUC value for conjoining the above two independent factors to predict grade 3 was 0.912 (95% CI 0.877-0.937, P＜0.001), and it was not significantly different from the AUC for using the mean CT value or CTR alone. Multivariate Cox regression analysis showed age (HR=1.05, 95% CI 1.02-1.09, P=0.003), CTR (HR=2.81, 95% CI 1.16-6.77, P=0.022), CT value (HR=2.49, 95% CI 2.30-15.43, P<0.001) and histological grade (HR=4.31, 95% CI 2.28-8.14, P＜0.001) were independent risk factors for DFS. Conclusion: Larger CTR and higher average CT value are independent predictors of higher IASLC histological grade. CTR (truncation values＜0.25 and ≥0.75) and average CT values (truncation values＜-410 HU and ≥-210 HU) can be used as preoperative substitutes for IASLC grading system.

Objective: To explore the ultrasound imaging evaluation of breast cancer microcalcifications under the expression status of different SOX gene family transcription factors. Methods: The study selected breast cancer patients diagnosed through pathological examination in Xingtai Central Hospital from January 2020 to October 2023. All patients underwent surgical treatment. Breast cancer tissue was obtained from the tumor site, along with adjacent normal tissue at least 5 cm away from the cancerous tissue. The collected tissues were immediately preserved in liquid nitrogen at -196℃ for subsequent research. Preoperative ultrasound examinations, including routine ultrasound, shear wave elastography, and contrast-enhanced ultrasound were performed. Postoperative pathological histological testing was conducted to detect SOX4 expression, and the relationship between ultrasonographic microcalcifications and SOX4 protein expression was analyzed. Results: In 215 patients, the proportion of high SOX4 expression in breast cancer tissue was 86.05%, significantly higher than that in adjacent normal tissue, which was 48.84% (χ²=67.783, P=0.000). Breast cancer microcalcifications were closely associated with clinical pathological features such as tumor differentiation and lymph node metastasis (P<0.05). High SOX4 expression in breast cancer was closely correlated with microcalcifications, Alder blood flow grading, peripheral radial enhancement, and the rim sign (P<0.05). Logistic regression analysis showed that microcalcifications (OR=1.839, 95% CI 1.538-2.198), peripheral radial enhancement (OR=1.795, 95% CI 1.089-2.959), and rim sign (OR=1.496, 95% CI 1.007-2.223) were risk factors for high SOX4 expression in breast cancer (P<0.05). Conclusion: SOX4 is implicated as a pivotal contributor to the pathogenesis and progression of breast cancer. The manifestation of microcalcifications in ultrasound imaging exhibits a significant correlation with tumor differentiation and lymph node metastasis, underscoring microcalcification as a prognostic marker for elevated SOX4 expression. Clinically, the identification of microcalcification patterns serves as a predictive tool for the expression of SOX4 transcription factors, which is anticipated to augment the precision of early diagnostic modalities and inform the development of targeted therapeutic strategies in breast cancer management.

Objective: To explore the value of multimodal ultrasound imaging in predicting central lymph node metastasis of papillary carcinoma of the thyroid (PTC). Methods: Patients with PTC confirmed by operation and pathology were selected in The First People’s Hospital of Foshan from March 2022 to February 2024. The patients were examined by conventional ultrasound, shear wave elastography (SWE) and superb microvascular imaging (SMI) before surgery. Conventional ultrasound was used to evaluate lesion maximum diameter, location, shape, internal parenchymal echo, microcalcification, the ratio of longitudinal and transverse diameter, capsule continuity. SWE was used to measure the mean of Young’s modulus (E_mean). SMI was used to measure vascular index (VI). Single-factor and multi-factor logistic regression analysis were used to analyze the risk factors of the central lymph node metastasis in PTC patients. Results: A total of 110 patients included 52 patients with central lymph node metastasis (metastatic group) and 58 patients without central lymph node metastasis (non-metastatic group). There were significant differences in age, maximum diameter, shape, parenchymal echo, microcalcification, capsule continuity, E_mean and VI between the two groups (all P<0.05). Multi-factor logistic regression analysis showed that age, maximum diameter, parenchymal echo and E_mean were the independent risk factors of central lymph node metastasis (all P<0.05). Conclusion: Multimodal ultrasound has a certain value in predicting the central lymph node metastasis of PTC.

Breast cancer is a highly heterogeneous malignancy, and molecular functional imaging techniques such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT) play a critical role in its accurate diagnosis. While imaging provides a crucial foundation for precise breast cancer diagnosis, the most used imaging agent,  18 F-FDG, presents certain limitations in this context. Consequently, there has been an increased focus on the research and development of novel radiolabeled probes for both in situ imaging of primary tumors and the detection of distant metastases. This paper reviewed the recent advancements in radiolabeled probes for breast cancer diagnosis and treatment, with an emphasis on novel targets that exploit specific surface markers on tumor cells and elements of the tumor microenvironment and categorized their practical applications in clinical settings.

Objective: To explore the application value of contrast-enhanced ultrasound in the diagnosis of Xp11.2 translocation /TFE3 gene fusion associated renal carcinoma (Xp11.2 translocation renal carcinoma). Methods: Patients with Xp11.2 translocation renal carcinoma confirmed by pathological examination were retrospectively enrolled as XP group between January 2021 and May 2023, while patients with clear cell renal cell carcinoma (ccRCC) during the same period were enrolled as ccRCC group. All patients underwent scans of routine ultrasound and contrast-enhanced ultrasound. The ultrasonographic findings in the two groups were compared. The diagnostic value of contrast-enhanced ultrasound in Xp11.2 translocation /TFE3 gene fusion-associated renal carcinoma was analyzed. Results: A total of 28 patients were included in the XP group and 40 in the ccRCC group. The age in XP group was younger than that in ccRCC group, and proportion of males was lower than that in ccRCC group (χ²=5.419; χ²=7.502, P<0.05). Routine ultrasound showed that there were significant differences in boundary and calcification between XP group and ccRCC group (P<0.05). Contrast-enhanced ultrasound showed that there were significant differences in perfusion mode and peak enhancement mode between XP group and ccRCC group (P<0.05). The time to peak (TTP) and begin to fade (BTF) in XP group were shorter than those in ccRCC group (P<0.05). There was no significant difference in arrival time (AT) between the two groups (P>0.05). Receiver operating characteristic (ROC) curve showed that AUC of TTP combined with BTF in the diagnosis of Xp11.2 translocation renal carcinoma was 0.855, greater than that of single index (0.773, 0.744; Z=2.202; Z=2.141, P<0.05). Binary logistic analysis showed that calcification, peak low enhancement or isoenhancement, TTP≤21.15 s, BTF≤25.31 s were all independent risk factors for the diagnosis of Xp11.2 translocative renal carcinoma by contrast ultrasound (P<0.05). Conclusion: Quantitative parameters of contrast-enhanced ultrasound have good differential diagnosis effect and clinical application value in Xp11.2 translocation renal carcinoma.

Breast cancer is highly heterogeneous at the molecular level, and molecular classification of breast cancer is very important for treatment selection and prognosis assessment. As a non-invasive and convenient diagnostic method, ultrasound plays an important role in breast cancer screening and diagnosis. By analyzing ultrasound image features and artificial intelligence modeling, medical personnel are expected to achieve non-invasive and accurate diagnosis of breast cancer molecular subtypes. This article reviewed the progress of ultrasonography in determining molecular typing of breast cancer.

Objective: To investigate the predictive value of a scoring model based on conventional ultrasound combined with clinicopathological features for pathologic complete response (pCR) in human epidermal growth factor receptor 2 (HER2) positive breast cancer after neoadjuvant chemotherapy (NAC). Methods: A retrospective analysis was performed in patients who confirmed as HER2 positive breast cancer by ultrasound-guided coarse needle biopsies and followed by eight cycles of NAC from January 2022 to August 2023 in The First Affiliated Hospital of Nanjing Medical University. According to the Miller-Payne grading system, the patients were divided into pathological complete response (pCR) group and non-pathological complete response (non-pCR) group. The differences of clinical and pathological data and ultrasound imaging features between the two groups were analyzed by t test or χ²/Fisher test. Multivariate regression analysis was used to determine the independent predictors of pCR in HER2 positive breast cancer, and a nomogram was established to visualize the predictive efficacy of related factors for pCR. Results: A total of 103 patients were included, 51 in the pCR group and 52 in the non-pCR group. Univariate analysis revealed significant differences between pCR and non-pCR groups regarding tumor molecular classification (χ²=12.266, P<0.001), as well as the ultrasound features including the rate of longest diameter change ΔD₂, ΔD₄, ΔD₆, ΔD₈ (t=-2.760, P=0.007; t=-2.557, P=0.012; t=-4.006, P<0.001; t=-2.872, P=0.005) and volume change rate ΔV₂, ΔV₄, ΔV₆, ΔV₈ (t=-4.167, P<0.001; Z=-3.443, P<0.001; Z=-4.518, P<0.001; Z=-3.708, P<0.001) along with resistance index (RI) measurements in the 4th, the 6th and the 8th cycle (Z=-2.108, P=0.035; Z=-2.386, P=0.017; Z=-3.708, P<0.001). Multivariate analysis showed that tumor molecular type (OR=0.15, 95% CI 0.03-0.64, P=0.005) and tumor volume change rate after the second cycle of NAC ΔV₂ (OR=121.47, 95% CI 4.25-3468.72, P=0.010) were independent predictors of pCR in HER2 positive breast cancer after NAC (P<0.05). The area under the receiver operating characteristic (ROC) curve (AUC) of the nomogram model was 0.82 (95% CI 0.74-0.91), the sensitivity was 85%, and the specificity was 75%. Calibration chart and decision curve analysis (DCA) showed it has good application value in clinical evaluation. Conclusion: The developed nomogram scoring model integrating ultrasound and clinicopathological features demonstrates strong predictive potential for identifying pCR among HER2 positive breast cancer patients post-NAC, thereby offering valuable insights into subsequent treatment decisions.

Objective: To compare the value of ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computer tomography (CT) and ⁶⁸GA-fibroblast activation protein inhibitor (FAPI) PET/CT in the diagnosis of different types of lymphoma. Methods: The clinical, pathological and imaging data of patients with lymphoma who received ¹⁸F-FDG and ⁶⁸Ga-FAPI PET/CT in Fudan University Shanghai Cancer Center within one week from November 2020 to October 2022 were retrospectively analyzed, and the detection rates of the two imaging agents for lesions were compared. At the same time, the differences of maximum standard uptake value (SUV_max), target-to-background ratio (TBR) and target-to-liver ratio (TLR) between the two groups were compared. Results: A total of 197 lymphomas (167 lymph node lesions and 30 extranodal lesions) were detected in 28 patients with pathologically confirmed lymphoma, and 191 lesions (161 lymph node lesions and 30 extranodal lesions) were detected by ¹⁸F-FDG PET/CT. ⁶⁸Ga-FAPI PET/CT detected 98 lesions (including 74 lymph node lesions and 24 extranodal lesions), and the detection rate of ¹⁸F-FDG PET/CT was significantly higher than that of ⁶⁸Ga-FAPI PET/CT. But for gastrointestinal infiltrating lesions, ¹⁸F-FDG PET/CT has a higher false positive rate. In lymphoma patients, SUV_max-FDG > SUV_max-FAPI, TBR-FDG > TBR-FAPI, the difference was statistically significant (14.4 vs 6.5, 13.3 vs 3.4, P<0.05). However, the SUV_max and TBR of Hodgkin’s lymphoma (HL) were 12.3 vs 6.5, 11.6 vs 5.1 (P>0.05) and indolent lymphoma 10.2 vs 10.9, 21.4 vs 5.1 (P>0.05), the difference was not statistically significant. There was no significant difference in TLR between the two examinations in all lymphoma patients (5.2 vs 5.2, P>0.05). Conclusion: The diagnostic value of ¹⁸F-FDG PET/CT for lymphoma is better than ⁶⁸Ga-FAPI PET/CT. But for HL and inert lymphoma, there was no significant difference in metabolic parameters between the two groups, and ⁶⁸Ga-FAPI PET/CT has higher diagnostic accuracy for gastrointestinal infiltration.

Objective: To evaluate the size and internal changes of bone metastases in breast cancer by spectral computed tomography (CT) images, to explore the predictive value of these multiple parameters for progression-free survival (PFS), and to overcome the limitations of conventional CT and existing criteria in the evaluation of bone metastases. Methods: The clinical and imaging data of patients with breast cancer with bone metastasis who were treated in Fudan University Shanghai Cancer Center from June 2019 to December 2022 were retrospectively analyzed. All patients underwent two contrast-enhanced spectral CT examinations, and the size indexes of bone metastases were recorded on the venous stage images: long diameter, short diameter, sum of long diameter and short diameter, and area of interest. Internal indicators of bone metastasis were CT value of conventional image (bone window) (CTv), CT value of single-level image (CT_{40 keV}, CT_{80 keV}, CT_{120 keV}), CT value of calcium suppression image (CT_CaSupp30, CT_CaSupp50, CT_CaSupp70, CT_CaSupp90). Each index of all the bone metastatic target lesions measured in the patient was summed respectively, and the change rate and monthly change rate of each index were calculated. The independent risk factors of PFS were determined by univariate and multifactor Cox proportional risk regression analysis combined with stepwise regression method. Cox proportional hazard regression model was used to construct the model, and the area under the time-dependent receiver operating characteristic curve (AUC) was used to evaluate its predictive performance. DeLong test compared AUC. The threshold was determined by X-tile method and divided into high and low risk groups according to the threshold. Kaplan-Meier (KM) survival curve and log-rank test were used to analyze the difference in prognosis between groups. Results: A total of 30 patients with breast cancer with bone metastases were included. The monthly rate of change of the sum of length and diameter of bone metastases and the monthly rate of change of CT_CaSupp70 are effective predictors of PFS in breast cancer patients with bone metastases. PFS was significantly prolonged in patients with the monthly change rate of the sum of length and diameter < 0.39% and the monthly change rate of CT_CaSupp70>-9.75%. The AUC of 1-year PFS predicted by the sum of long and short diameters model, CT_CaSupp70 model, combined model and response criteria of bone metastases at MD Anderson Cancer Center (MDA criteria) were 0.751, 0.702, 0.827 and 0.600, respectively. The predictive performance of the combined model was significantly better than that of the MDA standard (P=0.003). The PFS of high risk group were significantly lower than those of low risk group (P<0.05). There was no significant difference in PFS survival curve under MDA (P=0.091). Conclusion: The monthly rate of change in the sum of long and short diameters of bone metastases and the monthly rate of change in CT_CaSupp70 are effective predictive indicators for the prognosis of breast cancer patients with bone metastasis, and their combined application can enhance predictive performance.

Targeted radionuclide therapy (TRT) leveraging the unique properties of radioactive pharmaceuticals to concentrate therapeutic effects within tumor tissues. TRT can be principally categorized into two distinct approaches based on the differences in targeting mechanisms: chemical/biological targeting internal radiotherapy and physical targeting internal radiotherapy. Generally, TRT refers to the systemic administration of targeted radionuclide drugs. The continuous advancement of nuclear medicine diagnostic and therapeutic technologies, coupled with the advocacy of precision medicine, has led to an innovative integration in clinical practice: integrated physical targeted radionuclide theranostics. This article explored the current status and development of theranostics in physically targeted radionuclide treatment, highlighting some common and emerging therapies in clinical practice.

The application of immune checkpoint inhibitors (ICI) in cancer treatment is on the rise, conferring substantial survival benefits to patients. Nevertheless, the development of life-threatening myocarditis induced by ICI has emerged as a novel challenge in clinical practice. Multimodality imaging, due to their non-invasive and low-risk profile, play an important role in the evaluation of ICI-associated myocarditis. This article reviewed the advancements of multimodality imaging for the assessment of ICI-associated myocarditis.

During the therapeutic management of malignant neoplasms, for instance, during radiotherapy (RT) and chemotherapeutic regimens, treatment-associated cardiotoxicity may ensue, which is typically challenging to be detected and diagnosed at an early stage. Nuclear medicine molecular imaging is capable of directly revealing physiological and pathological processes at the cellular or molecular level, enabling the visualization of cardiotoxicity associated with the treatment of malignant tumors.This article provided a comprehensive review on the diagnostic significance of nuclear medicine molecular imaging in terms of cardiotoxicity associated with malignant neoplasm treatment from multiple perspectives, including myocardial metabolism, inflammation, neural injury, and fibrosis. The aim is to further elucidate its pathophysiological evolution and enhance the early-stage diagnostic proficiency.

Objective: To compare the diagnostic value of qualitative and quantitative parameters of sound touch elastography (STE) and strain elastography (SE) in differentiating benign and malignant Breast Imaging Reporting and Data System (BI-RADS) 4 breast lesions, and to explore the correlation between the parameters and the expression of human epidermal growth factor receptor 2 (HER2) in BI-RADS 4 breast malignant lesions. Methods: The clinicopathological data of patients with BI-RADS 4 breast lesions diagnosed by ultrasound examination in The First Hospital of Shanxi Medical University were selected. The quantitative and qualitative parameters of STE and SE were obtained: STE maximum elasticity (E_max), STE qualitative score, SE strain ratio (SR) and SE qualitative score. To compare the value of two ultrasonic elastography parameters in differentiating benign and malignant BI-RADS 4 breast lesions and their correlation with HER2 expression in BI-RADS 4 breast malignant lesions. Results: A total of 106 patients with 106 lesions were included. The difference of E_max, STE score, SR and SE score between benign and malignant lesions of BI-RADS 4 breast was statistically significant (P<0.05). Compared with the pathological results, the difference of E_max, STE score, SR and SE score in the diagnosis of benign and malignant BI-RADS 4 breast lesions according to the cut-off value was statistically significant (P<0.05). The diagnostic efficacy of E_max, STE score and SR was generally consistent with that of the gold standard (Kappa=0.436, 0.420, 0.502), and the diagnostic efficacy of SE score was more consistent with that of the gold standard (Kappa=0.626). The area under the receiver operating characteristic (ROC) curve (AUC) of SE score (AUC=0.883) was higher than that of E_max (AUC=0.743), STE score (AUC=0.756) and SR (AUC=0.783) diagnosis, with the greatest diagnostic efficiency, and the difference was statistically significant (P<0.05). In BI-RADS 4 breast malignant lesions, the difference of E_max and SE scores between HER2 positive expression and negative expression was statistically significant (P<0.05). E_max and SE scores were positively correlated with HER2 positive expression (r=0.458, 0.356). The best cut-off value of E_max and SE score in the diagnosis of HER2 positive expression was 123.75 kPa and 4 points, the sensitivity was 81.2% and 62.5%, the specificity was 60.5% and 68.4%, and the AUC was 0.766 and 0.689. Conclusion: The diagnostic efficacy of SE score was higher when STE and SE techniques were used to identify BI-RADS 4 breast lesions. E_max and SE scores were positively correlated with HER2 positive expression. E_max and SE score are of great value in identifying HER2 positive expression in BI-RADS 4 breast malignant lesions.

Hypoxia, a common status of most solid tumors, is associated with tumor progression, metastasis, radiotherapy and chemotherapy resistance. Positron emission tomography (PET) and single-photon emission computed tomography (SPECT) using hypoxic-targeting radiopharmaceuticals have been applied to non-invasive monitoring of tumor hypoxia. Nitroimidazole, which could be selectively trapped within hypoxia cells, has become an important pharmacophore for targeting hypoxia. Recently, hypoxic-targeting radiopharmaceuticals containing nitroimidazole have been developed for detecting tumor hypoxia. In this review, the progress of hypoxic-targeting radiopharmaceuticals containing nitroimidazole in recent ten years was introduced and the future development direction was also prospected.

Intrahepatic cholangiocarcinoma (ICC) is a highly aggressive liver malignancy with a poor prognosis, and early diagnosis is critical for improving patient survival. Conventional ultrasound has limited diagnostic value for ICC, and clinical diagnosis primarily relies on contrast-enhanced computed tomography (CECT) or contrast-enhanced magnetic resonance imaging (CEMRI). Contrast-enhanced ultrasound (CEUS), as a real-time, non-invasive, and convenient imaging technique, has demonstrated significant value in the diagnosis of focal liver lesion (FLL). This article systematically reviewed the current applications of CEUS in the diagnosis of ICC, with a particular focus on the controversies and latest advances in its development. Due to the overlapping imaging features of ICC and hepatocellular carcinoma (HCC) on CEUS, the diagnostic capability of CEUS was once questioned and even removed from the diagnostic tools for HCC in the guidelines of the American Association for the Study of Liver Diseases (AASLD). Subsequently, the academic community has engaged in extensive discussions on the value of CEUS in the differential diagnosis of ICC and HCC. Recent studies have shown that by optimizing the Liver Imaging Reporting and Data System (LI-RADS) criteria, incorporating dynamic quantitative analysis, and leveraging radiomics and artificial intelligence technologies, the accuracy and specificity of CEUS in diagnosing ICC have significantly improved. With the development of more high-quality studies, CEUS is expected to play an increasingly important role in the early diagnosis, classification, and treatment monitoring of ICC.

Objective: To assess the injuries of left ventricular function in breast cancer patients following anthracycline-based chemotherapy using ultrasonic vector flow mapping (VFM) combined with layer-specific two-dimensional speckle tracking imaging (LS2D-STI). Methods: A retrospective analysis was conducted on breast cancer patients who underwent chemotherapy with the TEC regimen (docetaxel 75 mg/m², epirubicin 90 mg/m², and cyclophosphamide 500 mg/m²) for the injuries of left ventricular function in Sichuan Provincial People’s Hospital from January to June 2017. A self-controlled study design was employed, with transthoracic echocardiography (TTE) performed before chemotherapy and at 48 h after each chemotherapy cycle, for a total of seven assessments (denoted as T0, T1, T2, T3, T4, T5, and T6) for parameters of intra-ventricular hemodynamic and transmural strain. Conventional echocardiographic parameters were collected, along with myocardial layer-specific global longitudinal strain, including endocardial global longitudinal strain (GLSendo), mid-myocardial global longitudinal strain (GLSmid), and epicardial global longitudinal strain (GLSepi). Additionally, left intra-ventricular hemodynamic parameters were measured, including global energy loss (EL), comprising early diastolic energy loss (EDC_EL), late diastolic energy loss (LDC_EL), isovolumic contraction energy loss (IVC_EL), and rapid ejection period energy loss (REP_EL). Left ventricular global circulation intensity (CIR) was also analyzed, including early diastolic circulation (EDC_CIR), late diastolic circulation (LDC_CIR), isovolumic contraction circulation (IVC_CIR), and rapid ejection period circulation (REP_CIR). The collected data were analyzed to assess changes in left ventricular function over the course of chemotherapy. Results: ① There were no significant differences in left ventricular ejection fraction (LVEF) between T1-T6 and T0 (P>0.05). The mitral inflow early diastolic peak velocity to mitral annular early diastolic peak velocity ratio (E/e’) gradually increased from T1 onward, with statistically significant differences observed (T0: 5.64 to T6: 8.76, P<0.05). ② There were no significant differences in GLSendo, GLSmid, or GLSepi at any time point compared to baseline (P>0.05). However, these parameters showed a trend of initially increasing and then decreasing, with strains at T6 being lower than baseline values. ③ EL showed an overall downward trend, while CIR exhibited an upward trend. Specifically, LDC_EL at T2-T6 was significantly lower than T0, whereas LDC_CIR at T2, T4, and T6 was higher than T0. EDC_EL and IVC_EL were significantly lower at T5 and T6, and REP_EL at T5 and T6 was significantly lower than baseline (P<0.05). Changes in LDC-phase parameters occurred earlier than those in other phases. EL and CIR changes occurred later than changes in E/e’. Conclusion: Ultrasonic VFM and LS2D-STI contribute to comprehensive assessment of cardiotoxicity in breast cancer patients undergoing anthracycline-based chemotherapy. Among these, the combination of E/e’ with EL and CIR may have more significant clinical value in evaluating changes in left ventricular diastolic function.

Objective: To evaluate the protective effects of astragaloside Ⅳ (AS-Ⅳ) against epirubicin (Epb)-induced cardiac injury in rats using speckle tracking imaging. Methods: Thirty-two SD rats were randomly assigned to four groups (n=8 per group): normal control, Epb model, low-dose AS-Ⅳ, and high-dose AS-Ⅳ. Transthoracic echocardiography (TTE) was performed at baseline and at 2, 4, and 6 weeks post-treatment to assess left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS). At week 6, serum levels of CK-MB, LDH, and IL-17 were measured, and heart tissues were collected for Masson staining to evaluate myocardial fibrosis. Results: LVEF showed no significant changes across groups or over time (P>0.05). GLS, however, was significantly influenced by both factors [F group (3, 21)=11.014 2, P<0.000 1] and time [F time (3, 21)=7.215 4, P=0.001 7], with a near-significant interaction effect [F interaction (9, 63)=1.876 1, P=0.072 0]. Serum CK-MB and LDH levels showed no significant differences among groups (P>0.05). IL-17 levels were significantly elevated in the Epb model group compared to the control group, while both AS-Ⅳ treatment groups exhibited significantly reduced levels compared to the Epb model group (P<0.05). The heart weight index did not differ significantly among groups, but myocardial fibrosis area showed significant variations, worsening progressively from control to high-dose AS-Ⅳ, low-dose AS-Ⅳ and Epb model groups（P<0.000 1）. Conclusion: AS-Ⅳ may attenuate myocardial fibrosis and reduce inflammation in Epb-induced myocardial damage. GLS proved to be a more sensitive and accurate indicator than LVEF for detecting subclinical myocardial injury.

 Objective: To explore the application value of radiomics model based on multi-parameter magnetic resonance imaging (mpMRI) in Gleason grading of prostate cancer. Methods: The data of patients who underwent prostate mpMRI examination with surgical or pathological puncture results confirming prostate cancer at Ma’anshan People’s Hospital from November 2020 to August 2023 were retrospectively analyzed. MpMRI data were extracted, including T2- weighted imaging (T2WI), zoomed imaging technique with parallel transmission diffusion-weighted imaging (ZOOMit DWI) and apparent diffusion coefficients (ADC). Spearman’s correlation coefficient was used to preliminarily screen the histological features, the least absolute shrinkage and selection operator (LASSO) algorithm and ten-fold cross-validation were used to further screen, logistic regression was used to construct the model, and the receiver operating characteristic (ROC) curve was used to judge the results. And the area under the ROC curve (AUC) was compared between models using the DeLong test. Results: A total of 176 patients were included, including 72 patients in the low-grade group (Gleason score≤3+4) and 104 patients in the high-grade group (Gleason score≥4+3), who werer andomly divided into training group (n=141) and test group (n=35) according to 7∶3. A variety of classifiers were used to construct the multi-parameter model, and the results showed that the AUC of support vector machine (SVM) in the test set was 0.891, and the AUC in the training set was 0.905. Light gradient boosting machine (LightGBM) had the highest AUC of 0.931 in the training set, but it performed poorly in the test set with an AUC of 0.808. The AUCs of multilayer perceptron (MLP) in the test set and the training set were 0.883 and 0.855, respectively, which were weaker than that of SVM, which showed that the stability of LightGBM and MLP models were slightly worse than that of SVM. In addition, the overall performance of the four methods ［k-nearest neighbor (KNN), extra trees (ET), random forest (RF), extreme gradient boosting (XGBoost)］were not as good as SVM, and some of them are overfitted. In general, in terms of Gleason grading of prostate cancer, the SVM model had a higher AUC in both the test set and the training set, and its stability and model classification ability were better. Conclusion: Constructing a multimodal imaging histology model based on mpMRI has significant clinical application value in Gleason grading of prostate cancer, of which the SVM model is the best.

Objective: To evaluate the performance of a deep learning model based on the Kupffer phase of perflubutane  microspheres for injection (product name Sonazoid) contrast-enhanced ultrasound in predicting microvascular invasion (MVI) of hepatocellular carcinoma (HCC), comparing it with radiomics model and clinical model. Methods: This study retrospective included 146 patients with primary HCC who underwent Sonazoid contrast-enhanced ultrasound examination in The First Affiliated Hospital of Guangxi Medical University from July 2020 to September 2022, randomly divided into a training set of 102 and a validation set of 44 in a 7∶3 ratios. Based on the region of interest in tumors, ResNet101 model was used to extract deep learning features through transfer learning, and PyRadiomics was utilized to extract radiomics features. Mann-Whitney U test and least absolute shrinkage and selection operator (LASSO) algorithm were employed to reduce features dimension. LASSO regression was used to construct both the deep learning model and radiomics model, a clinical model was also built based on clinical features. The diagnostic performance of models was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. DeLong testing algorithm was used to compare the diagnostic performance between models. Results: In the training set, the AUC (95% CI) for the deep learning model, radiomics model, clinical model was 0.931 (0.880-0.981), 0.823 (0.744-0.903) and 0.719 (0.614-0.824), respectively. In the validation set, the
AUC (95% CI) for the deep learning model, radiomics model, clinical model was 0.895 (0.757-1.000), 0.711 (0.514-0.909) and 0.606 (0.390-0.822), respectively. DeLong testing indicated that in both the training and validation sets, the diagnostic performance of the deep learning model was superior to that of the radiomics model and clinical model (P＜0.05). Both univariate and multivariate logistic regression analyses showed that AFP (P＜0.05) and Barcelona Clinic Liver Cancer staging (P＜0.001) could be used as independent predictors of MVI in HCC patients. Conclusion: The deep learning model based on the Kupffer phase of Sonazoid contrast-enhanced ultrasound demonstrates excellent performance in predicting MVI in HCC patients. It has the potential to become a non-invasive imaging biomarker for predicting MVI.

Objective: To explore the potential value of multiparametric magnetic resonance imaging (mpMRI) radiomics combined with machine learning models in predicting sentinel lymph node (SLN) metastasis in breast cancer. Methods: A retrospective analysis of imaging and clinical data from pathologically confirmed breast cancer patients was conducted, and the patients were divided into training and validation groups in a 7∶3 ratio. Radiomics features were extracted from T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC), and the second phase of dynamic contrast-enhanced MRI (DCE-MRI), including the whole tumor (ROI_Whole) and subregions (ROI_Sub). Multiple machine learning models were constructed by integrating clinical, pathological, and imaging features, and their predictive performance was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis, and calibration curves. Results: This study analyzed 193 breast cancer patients, including 80 SLN-positive and 113 SLN-negative cases. Univariate and multivariate logistic regression analyses were performed on clinical, imaging, and pathological data to identify independent risk factors associated with SLN metastasis. Finally, it is analyzed that，peritumoral edema and lymphovascular invasion (LVI) showed significant differences between SLN-positive and SLN-negative groups (P<0.01), with LVI identified as an independent predictive factor (P<0.05). The MLP model, incorporating radiomics features from T2WI, ADC, ROI-Whole, ROI-Sub, and LVI, demonstrated the best performance, with an AUC of 0.947 in the training set and 0.932 in the validation set. Conclusion: The radiomics model based on breast mpMRI effectively predicts SLN metastasis in breast cancer preoperatively.

Objective: To utilize the advantages of digital breast tomosynthesis (DBT) in assessing lesion margins and to explore the relationship between the burr sign of DBT images and Ki-67 proliferation index. Methods: DBT imaging data of 99 patients with invasive breast cancer who in the First Affiliated Hospital of Zhengzhou University from March 2022 to April 2023 were retrospectively included, and all of the patients showed a burr-type mass in DBT images. Lump size, length and width of the burr, coverage of the burr at the tumor margin, and number of burrs were analyzed in 99 cases of breast burr-type lumps, and general clinical data of the patients were collected to compare the differences of each parameter between the Ki-67 proliferation index expression states. Independent predictors of Ki-67 proliferation index were analyzed using multifactorial logistic regression, and the diagnostic efficacy was evaluated using subject working curves. Results: The differences in DBT image burr characteristics including burr length and burr width were statistically significant when comparing Ki-67 proliferation index high patients and low patients (P＜0.05), whereas the differences in the number of burrs, age of patients, menopausal status, and size of the mass were not statistically significant (P=0.060, P=0.175, P=0.507, and P=0.050, respectively). Multifactorial logistic regression model analysis showed that burr length (OR=0.036, P＜0.001) and burr width (OR=8.829, P＜0.001) were independent predictors of Ki-67 proliferation index. The best diagnostic efficacy was achieved when combining burr length with burr width, with an AUC of 0.897. Conclusion: Burr sign analysis in DBT images of breast cancer can be used as a noninvasive predictor of the proliferative activity of malignant tumors to determine patient prognosis.

 In recent years, artificial intelligence (AI) technology has experienced a rapid development in the field of medical imaging, with increasing applications in bone tumor imaging. These applications encompass tasks such as tumor segmentation, classification, chemotherapy efficacy evaluation, and prognostic prediction. These advancements facilitate the development of related computer-aided diagnostic tools, which assist in reducing misdiagnoses and unnecessary biopsies. Furthermore, they can guide the formulation of clinical decisions and personalized treatment strategies, thereby advancing the process of precise treatment for bone tumors. Multiple studies have established the superiority of AI technology. With the gradual expansion of sample sizes and continuous refinement of models, AI models are expected to gradually be integrated into clinical practice and emerge as essential tools in the diagnosis and treatment of bone tumors, aiding precision medicine.