Objective To explore the value of a deep learning-based ultrasound radiomics nomogram in predicting Ki-67 expression levels in invasive breast cancer. Methods A retrospective single-center study was conducted, collecting complete preoperative clinical data and ultrasound images from 465 patients with pathologically confirmed invasive breast cancer at the First Affiliated Hospital of Bengbu Medical University from January to December 2024. Image acquisition was performed using Mindray Resona 7 and Samsung HS60 color Doppler ultrasound systems. Based on immunohistochemical results, patients were divided into high and low Ki-67 expression groups and randomly assigned to training (n=326) and validation (n=139) cohorts at a 7:3 ratio. ITK-SNAP software was used to segment tumors from the largest 2D ultrasound cross-sectional images, with interobserver consistency of ROI delineation assessed by ICC. Pyradiomics was employed to extract radiomics features from tumor tissues, and four deep learning networks were pretrained to construct clinical, ultrasound radiomics, fusion, and combined nomogram models. Diagnostic performance and clinical utility were evaluated using ROC curves, calibration curves, and decision curve analysis. Results Nineteen optimal ultrasound radiomics features and the DenseNet121 deep learning model showed the best performance (P<0.05). In the training cohort, the AUCs for the clinical model, ultrasound radiomics model, deep learning model, fusion model, and nomogram were 0.79 (95% CI: 0.74-0.84), 0.85 (95% CI: 0.81-0.90), 0.87(95% CI: 0.83-0.91), 0.94(95% CI: 0.91-0.97), and 0.95(95% CI: 0.93-0.98), respectively. In the validation cohort, the corresponding AUCs were 0.76(95% CI: 0.68-0.84), 0.78 (95% CI: 0.70-0.85), 0.81(95% CI: 0.74-0.88), 0.91(95% CI: 0.86-0.96), and 0.93(95% CI: 0.89-0.98). Conclusion The deep learning-based ultrasound radiomics nomogram can effectively predict Ki-67 expression in invasive breast cancer.

Objective To compare the image quality of non-contrast MR pulmonary angiography (MRPA) with two different voxel sizes. Methods This study consisted of 29 healthy volunteers recruited in our hospital from Sptember to October 2020 who underwent non-contrast MRPA with the following two sequences of different parameters: Group 1: voxel size=1.2 mm×1.2 mm×4 mm, group 2: voxel size=2 mm×2 mm×2 mm. The subjective image quality assessment of normal pulmonary arteries was evaluated by two experienced radiologists on primary coronal images and constructed axial images separately and image quality of these two groups was compared. Results The mean image quality scores of the main, left, and right pulmonary arterial trunks on primary coronal images were 3(3,3), 3(3,3), 3(3,4) (group 1) and 2(2,3), 2(2,3), 2(3,3) (group 2). The mean image quality scores of the main, left, and right pulmonary arterial trunks on constructed axial images were 2(2,3), 1(1,2), 1(2,2) (group 1) and 1.5(2,2), 2(2,3), 2(2,3) (group 2). Branch subjective image quality of group 1 was better than group 2 on primary coronal images (P<0.05), however, there was no statistical difference between these two groups in signal noise ratio (SNR) and contrast noise ratio (CNR, P>0.05). Meanwhile, the subjective image quality of the left and right pulmonary artery trunk in group 2 was better than group 1 on constructed axial images (P<0.05), in contrast, SNR and CNR of all pulmonary artery branches evaluated in group 1 was better than group 2 on constructed axial images (P<0.05). Otherwise, SNR and CNR of all pulmonary artery branches evaluated on coronal images were better than constructed axial images (P<0.05). Conclusion Non-contrast MRPA images with higher resolution within a coronal plane with voxel size 1.2 mm×1.2 mm×4 mm can provide better primary coronal images but images with isotropy with voxel size 2 mm×2 mm×2 mm can acquire better subjective image quality on constructed axial images.

Objective To explore the feasibility of ultra-high frequency ultrasound to examine the extracranial segment of facial nerve, to describe the normal ultrasonic appearance and measurements of extracranial segment of facial nerve, and to study the influencing factors. Methods Ultra-high frequency ultrasound was used to explore the bilateral extracranial facial nerves of a total of 70 healthy volunteers recruited at West China Hospital of Sichuan University from February 10th to March 1st, 2023. The consistency of the measurements was studied, and the differences in diameter, circumference, and visible length of the facial nerve at different sites between the left and right sides were compared. The effects of gender, age, BMI, occupation related daily voice use, and measurement sites on the facial nerve were also investigated. Results The inter group and intra group correlation coefficient values measured by ultrasound of the extracranial segment of the facial nerve are all greater than 0.75. There were no statistically significant differences in the diameter, circumference, and visible length of the bilateral extracranial segments of the facial nerve (P>0.05). These measurements were not significantly affected by the measurement sites, gender, age, or BMI (P>0.05). However, the circumference of the facial nerve was lower in the group with shorter daily voice use (P<0.05). Conclusion There is good consistency in ultra-high frequency ultrasound measurements of the diameter, circumference, and length of the extracranial segment of the facial nerve, and some of these measurements are influenced by occupation related daily voice use.

Objective To investigate the diagnostic value of 3.0T MRI T2 mapping combined with diffusion tensor imaging (DTI) in peripheral neuropathy associated with type 2 diabetes mellitus (DPN). Methods A prospective cohort of 42 patients with DPN hospitalized from January 2021 to April 2024 (DPN group), along with 29 healthy controls (HC group), was enrolled. All participants underwent lower limb nerve T2 mapping, DTI, and whole-body diffusion-weighted imaging with background suppression on a 3.0T MRI system. Quantitative parameters were measured, including T2 values of the sciatic and tibial nerves, DTI metrics, quadriceps T2 values, the T2 ratio of sciatic nerve to quadriceps at the same level, and cross-sectional areas of the sciatic and tibial nerves. Group differences were assessed, and diagnostic performance was evaluated using ROC curve analysis. Results Compared with the HC group, patients with DPN exhibited significantly increased T2, axial diffusivity (AD), and radial diffusivity (RD) values of the sciatic and tibial nerves (P<0.001), while fractional anisotropy (FA) and relative anisotropy (RA) values were decreased (P<0.001). Within the DPN group, significant differences were observed between the sciatic and tibial nerves in T2, FA, RD, and RA values (P<0.05), whereas AD values showed no significant difference (P>0.05). No significant intra-group differences were found in the HC group (P>0.05). The quadriceps T2 values and sciatic nerve-to-muscle T2 ratios were higher in the DPN group compared to HC (P<0.05). The cross-sectional areas of both sciatic and tibial nerves also differed significantly between groups (P<0.05). ROC curve analysis revealed that the AUC of T2 values for the sciatic and tibial nerves was 0.864 and 0.726, respectively; for FA values, the AUC was 0.825 and 0.800; and for combined T2 and FA values, the AUC was 0.895 and 0.822, respectively. Conclusion 3.0T MRI T2 mapping combined with DTI enables effective quantitative assessment of diabetic peripheral neuropathy. A diagnostic model integrating T2 and FA values provides superior diagnostic accuracy. This approach not only reveals fat infiltration and water content changes in the quadriceps, but also evaluates morphological alterations of lower limb nerves in DPN patients, offering reliable imaging evidence for early diagnosis and disease assessment.

Objective To evaluate the relationship between ultrasound-derived fat fraction (UDFF) and non-contrast multislice helieal computed tomography (MSCT) CT values, aiming to estimate hepatic fat content using UDFF. Methods A total of 140 patients from Beijing Tsinghua Changgung Hospital were prospectively enrolled between February 2024 AND June 2024. Based on the order of admission, the first 110 patients were assigned to the training set and the subsequent 30 patients to the validation set. All patients underwent both ultrasonography and upper abdominal MSCT within 7 days. The training set was categorized into a normal group (hepatic fat content<5.0%, n=68), a mild fatty liver group (hepatic fat content≥5.0% and <10.0%, n=30), and a moderate-to-severe fatty liver group (hepatic fat content≥10.0%, n=12). UDFF and CT values were measured in the right hepatic lobe using regions of interest (ROI) analysis. Correlation between UDFF and CT values was assessed, and comparisons were made among different fatty liver groups. Diagnostic performance of UDFF was evaluated using ROC curve analysis. Results In the training set, UDFF showed a strong negative correlation with CT values (r=-0.738, P<0.001), with statistically significant differences among groups (P<0.001). The UDFF values in the moderate-to-severe fatty liver group were significantly higher than those in the mild fatty liver and normal groups and normal groups (P<0.001). For diagnosing mild fatty liver, the area under the curve (AUC) was 0.661 (P=0.011、95% CI: 0.536~0.785), with a sensitivity of 43.3%, specificity of 85.3%, and a cutoff value of 6.1%. For moderate-to-severe fatty liver, the AUC was 0.921 (P<0.001、95% CI: 0.809~1.032), with a sensitivity of 91.7%, specificity of 93.3%, and a cutoff value of 10.4%. In the validation set, UDFF demonstrated an accuracy of 71.4%, sensitivity of 57.1%, specificity of 76.2% for mild fatty liver diagnosis, and an accuracy of 88.9%, sensitivity of 100%, specificity of 85.7% for moderate-to-severe fatty liver diagnosis. Conclusion UDFF exhibits a strong negative correlation with CT values in the quantitative diagnosis of fatty liver, and demonstrates high accuracy in diagnosing moderate-to-severe fatty liver.

Objective To develop a deep learning-based super-resolution reconstruction framework for small hepatocellular carcinoma (sHCC) in magnetic resonance imaging,designed to synergistically enhance lesion conspicuity and diagnostic utility by jointly optimizing high-frequency detail preservation and anatomical fidelity in thin-slice diffusion-weighted imaging (DWI). Methods A retrospective study was conducted on 3-mm DWI data from 300 patients with sHCC admitted from December 2022 to June 2024. The dataset was randomly divided into training and test sets at an 8:2 ratio. A dual-branch super-resolution model was constructed, comprising a content branch for extracting global features through cascaded gradient Transformer blocks and a gradient branch for enhancing structural information via gradient Transformer blocks. The model incorporated two novel components: (1) a cross local-enhanced self-attention module to optimize interactions between pixel-level features and global context, and (2) a channel-spatial joint attention layer that dynamically allocates weights to improve the visibility of key anatomical structures. Three senior radiologists blindly evaluated both original (OR-DWI) and super-resolved (SR-DWI) images using a 5-point Likert scale. Objective metrics, including the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), were also calculated. Results SRDWI demonstrated significantly superior subjective ratings compared to ORDWI across multiple evaluation metrics? (P<0.001), including hepatic nodule signal specificity (3.41±0.53 vs 2.47±0.50), normal liver parenchyma homogeneity (3.29±0.47 vs 2.78±0.42), artifact interference severity (2.56±0.52 vs 2.47±0.48), and overall image quality (3.15±0.49 vs 2.67±0.48). Although post-super-resolution processing resulted in a marginally increased image noise level (3.98±0.61 vs 2.87±0.46, P<0.05), the values remained well within clinically acceptable limits without compromising diagnostic interpretation of key features. Quantitative analysis further confirmed SRDWI's excellent performance, achieving a peak signal-to-noise ratio of 34.65489 dB and structural similarity index of 0.90365. Conclusion The proposed deep learning framework effectively bridges the diagnostic performance gap between thin-slice (3 mm) and conventional thick-slice DWI protocols, achieving comparable lesion characterization accuracy while mitigating motion artifacts. This advancement establishes a robust technical foundation for precision diagnosis of early-stage sHCC, where subtle anatomical details critically influence therapeutic decision-making.

Objective To investigate the apllication value of deep learning image reconstruction (DLIR) algorithms in optimising the images quality of low-kV head and neck CT angiography (CTA). Methods Sixty patients who underwent head and neck CTA in the Affiliated Hospital of Shaanxi University of Traditional Chinese Medicine Hospital from October 2024 to February 2025 were analyzed. The scanning was performed with 80 kV and smart mA (50-400 mA). The contrast agent dosage was 40 mL, the flow rate was 4 mL/s, and the injection time of contrast agent was 10 s. After scanning, 40% adaptive iterative reconstruction algorithm-V (ASIR-V40%), low intensity DLIR (DLIR-L), medium intensity DLIR (DLIR-M) and high intensity DLIR (DLIR-H) were reconstructed respectively. The CT and SD values of the aortic arch, 2 cm above the opening of the common carotid arteries bilaterally, segment V1 of the vertebral arteries, proximal basilar artery, segment M1 of the middle cerebral artery, and temporalis muscle were measured, and signal-to-noise ratios (SNR) and contrast-to-noise ratios (CNR) were calculated. The images were subjectively scored by two radiologists using a five-point scale in a double-blind method. The SD value, CT value, CNR value, SNR value and subjective score of the four reconstruction algorithms were compared. Results There was no statistically significant difference in CT values for each vessel between the four reconstruction algorithms (P>0.05). As the reconstruction levels of ASIR-V40%, DLIR-L, DLIR-M and DLIR-H gradually increased, the images showed a decreasing trend in SD, and an increasing trend in CNR and SNR.Among them, DLIR-H has the lowest SD value and the highest CNR and SNR. Moreover, there are statistically significant differences in the SD, SNR, and CNR values between DLIR-L, DLIR-M, DLIR-H and ASIR-V40% (P<0.05).Compared with ASIR-V40%, DLIR-H Reconstruction the SNR and CNR of the aortic arch, 2 cm above the common carotid artery opening, V1 segment of the vertebral artery, proximal basilar artery, and M1 segment of the middle cerebral artery in DLIR-H increased by approximately 69%, 48.3%, 48.3%, 51.6%, 56.4% and 68%, 55.9%, 48.2%, 51.6%, 56.6% respectively, and the SD values decreased by approximately 40.2%, 35.8%, 30.6%, 33.2%, 35.3% respectively.The subjective scores of DLIR-H, DLIR-M and DLIR-L were all higher than those of ASIR-V40%, and the differences were statistically significant (P<0.05). Conclusion Compared with ASIR-V at 40%, DLIR reconstruction can further reduce image noise and improve image quality in head and neck CTA. Therefore, the DLIR algorithm can be used to enhance the image quality of low-kV head and neck CTA.

Objective To investigate the value of combining contrast-enhanced ultrasound (CEUS) features with quantitative parameters from spectral CT in predicting the World Health Organization/International Society of Urological Pathology (WHO/ISUP) grading of clear cell renal cell carcinoma (ccRCC). Methods A retrospective analysis was conducted on 115 patients with pathologically confirmed ccRCC who underwent surgery at the First Hospital of Shanxi Medical University from June 2019 to September 2024. Based on WHO/ISUP grading, patients were classified into a low-grade group (grades 1-2, n=79) and a high-grade group (grades 3-4, n=36). All patients underwent preoperative CEUS and spectral CT examinations. Univariate analysis was used to compare CEUS features and spectral CT quantitative parameters between groups. Statistically significant parameters were selected to construct logistic regression models. The predictive performance of each model was assessed by calculating the area under the ROC curve (AUC). Results Univariate analysis revealed that several CEUS features, including contrast agent perfusion pattern, enhancement homogeneity, perilesional rim-like hyperenhancement, lesion boundary clarity, lesion morphology, and lesion extent before and after enhancement, these were significantly different between the two groups (P<0.05). Spectral CT parameters showing significant differences included lesion maximum diameter, corticomedullary phase CT value, iodine concentration, and normalized iodine concentration (P<0.05). Multivariate logistic regression analysis demonstrated that the CEUS model and the spectral CT model achieved sensitivities of 0.639 and 0.889, specificities of 0.848 and 0.785, accuracies of 0.774 and 0.809, and AUCs of 0.797 and 0.874, respectively. The combined model achieved superior diagnostic performance, with a sensitivity of 0.917, specificity of 0.810, accuracy of 0.852, and an AUC of 0.893. Conclusion The combination of CEUS features and spectral CT quantitative parameters provides a more accurate prediction of WHO/ISUP grading in ccRCC, offering valuable support for precision treatment in clinical practice.

Objective To identify the model with the highest diagnostic value for differentiating ovarian cystadenoma (OCA) from ovarian cystadenocarcinoma (OCAC) by extracting and selecting features from MRI imaging data to establish radiomics and traditional MRI diagnostic models. Methods A retrospective analysis was conducted on 173 patients (82 OCA cases and 91 OCAC cases) confirmed by pathology and preoperatively undergoing contrast-enhanced MRI scans at the First Affiliated Hospital of Bengbu Medical University from January 2020 to December 2024. Univariate and multivariate regression analyses were performed on traditional MRI features to identify diagnostic predictors for OCA and OCAC, and diagnostic models were developed and evaluated. Patients were randomly divided into training (n=121) and testing (n=52) cohorts at a 7:3 ratio. Regions of interest (ROI) were delineated on T1-weighted contrast-enhanced (T1WI-CE) and T2-weighted (T2WI) sequences using the United Imaging Intelligence Research Platform. Extracted radiomic features underwent max-min normalization, Select K Best, and Least Absolute Shrinkage and Selection Operator regression for dimensionality reduction and optimal feature selection. Logistic regression models were constructed for T1WI-CE, T2WI, and combined T1WI-CE+T2WI. Diagnostic performance was evaluated using the AUC, calibration curves, and decision curve analysis (DCA). Results The combined T1WI-CE+T2WI radiomics model demonstrated superior diagnostic efficacy (AUC=0.886) compared to individual T1WI-CE, T2WI, and traditional MRI models. Conclusion MRI radiomics-based diagnostic models can effectively differentiate OCA from OCAC, providing guidance for clinical treatment strategies.

Objective To investigate the value of a radiomics model integrating multimodal MRI and mammography (MG) for predicting axillary lymph node (ALN) metastasis in breast cancer. Methods A retrospective study was conducted, enrolling 575 breast cancer patients from The Xuzhou Central Hospital and The Huai'an First People's Hospital from January 2021 to May 2024. By comparing clinical indicators between the ALN metastasis-positive and negative groups, two clinical factors significantly associated with ALN metastasis (P<0.05) were identified: clinical T stage and lymph node palpation findings. Based on lesion segmentation from dynamic contrast-enhanced MRI (DCE-MRI) and MG images, radiomic features were extracted and screened, retaining 24 features with non-zero coefficients. Three Random Forest models were constructed based on features from the MRI tumor region (MRI_Tumor), the MRI lymph node region (MRI_LN), and the MG tumor region (MG_Tumor). These three models were then combined to form a clinical factor-free radiomics model (All_Imaging_Fusion). Finally, by integrating clinical T stage and lymph node palpation, a comprehensive multimodal radiomics model (All_Imaging+Clinical_Fusion) was developed. Results The multimodal radiomics model (All_Imaging+Clinical_Fusion) demonstrated excellent predictive performance, with AUC values of 0.943, 0.931, and 0.911 in the training, internal validation, and external validation cohorts, respectively. It significantly outperformed models based on any single imaging modality (P<0.05). Other predictive efficacy metrics also confirmed its superiority. Decision curve analysis indicated that this model provides a high clinical net benefit. Conclusion The multimodal radiomics model, which integrates radiomic features from DCE-MRI and MG with key clinical factors, can non-invasively and effectively predict ALN metastasis in breast cancer, offering a novel approach for guiding individualized treatment strategies.

Objective To analyze the MRI features and risk factors of recurrent anal fistula. Methods The clinical data of 120 patients with anal fistula who underwent surgical treatment at the First Affiliated Hospital of Anhui University of Chinese Medicine from May 2016 to May 2025 were retrospectively analyzed. According to the occurrence of anal fistula, the patients were divided into the recurrent group (n=40) and the newly diagnosed group (n=80). MRI features and general data of the two groups were comparatively analyzed. Multivariate logistic regression was used to analyze the influencing factors of recurrent anal fistula. Results The proportions of patients with high anal fistula, number of fistulas>1, course of disease>1 year, and history of anorectal surgery in the recurrent group were higher than those in the newly diagnosed group (P<0.05). Multivariate logistic regression analysis showed that high anal fistula (OR=1.471, 95% CI: 1.177-1.839), number of fistulas>1 (OR=1.527, 95% CI: 1.209-1.928), course of disease>1 year (OR=1.508, 95% CI: 1.221-1.864), and history of anorectal surgery (OR=1.496, 95% CI: 1.201-1.864) were risk factors for recurrent anal fistula (P<0.05). Conclusion Recurrent anal fistula exhibits certain features on MRI. The location of anal fistula, number of fistulas, course of disease, and history of anorectal surgery are factors influencing recurrent anal fistula. Targeted prevention and control measures should be developed based on above factors to reduce postoperative recurrence risk.

Objective To evaluate the value of deep learning reconstruction (SupMR) in optimizing scan time, signal-to-noise ratio (SNR), and structural similarity for head diffusion-weighted imaging (DWI). Methods A total of 40 healthy volunteers were prospectively and consecutively enrolled from December 2024 to January 2025 in Affiliated Hospital of Yangzhou University. All participants underwent head MRI scans using both low number of excitations diffusion-weighted imaging (LN-DWI) and conventional DWI (C-DWI) sequences. The LN-DWI images were post-processed using SupMR to generate Sup-DWI images, with the scan time recorded for each sequence. Three sets of images (LN-DWI, C-DWI, and Sup-DWI) were evaluated by subjective and objective methods (region-of-interest signal-to-noise ratio, contrast-to-noise ratio (CNR), apparent diffusion coefficient (ADC), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Intraclass correlation coefficient (ICC) was used to analyze the consistency of ADC values among the three groups, as well as the intra- and inter-observer agreement of subjective scores between two radiologists. Results Compared with the C-DWI group, the scanning time of Sup-DWI group was reduced by 52%. The results of consistency analysis showed good intra- and inter-observer consistency between the two radiologists (ICC>0.75), and good consistency of ADC values in all brain parenchyma regions between each pair of groups (ICC>0.75). The results of subjective score analysis showed a statistically significant difference among the three groups (P<0.05); pairwise comparison showed no significant difference between Sup-DWI group and C-DWI group (P>0.05). Objective evaluation and analysis showed that the differences in SNR and CNR between the three groups were statistically significant (P<0.05); pairwise comparison showed that the differences in SNR and CNR were statistically significant (P<0.05). The PSNR and SSIM values of the Sup-DWI group were greater than those of the LN-DWI group, and the differences were statistically significant (P<0.05). Conclusion SupMR technology can significantly shorten the DWI scanning time, optimize image quality, and help improve the efficiency of examination.

Objective To investigate the prognostic value of chest CT combined with serological indexes [hypersensitive C-reactive protein (hs-CRP), D-dimer (D-D), procalcitonin (PCT)] in patients with severe pneumonia. Methods A total of 80 patients with severe pneumonia and 60 patients with common pneumonia in the Sixth People's Hospital of Nantong were enrolled as severe group and common group from November 2020 to August 2024, respectively. All patients underwent chest CT plain scan, and levels of hs-CRP, D-D and PCT were detected. According to 28d survival, patients in severe group were divided into death group (n=29) and survival group (n=51). CT scores, hs-CRP, D-D and PCT in different groups were compared, and their prognostic value in severe patients was evaluated. Results CT score, levels of hs-CRP, D-D and PCT in severe group were higher than those in common group (P<0.05). APACHE II score at admission, proportions of infection strains≥2 kinds, disorder of consciousness, organ dysfunction≥3, septic shock and mechanical ventilation, CT score, levels of hs-CRP, D-D and PCT in death group were higher than those in survival group (P<0.05). High APACHE II score at admission, organ dysfunction ≥3 and high CT score were independent risk factors of 28 d prognosis (death) in patients with severe pneumonia (P<0.05). The area under the curve (AUC) values of CT score, hs-CRP, D-D and PCT for predicting prognosis of patients with severe pneumonia were 0.714 (95% CI: 0.592-0.836), 0.686 (95% CI: 0.564-0.808), 0.727 (95% CI: 0.607-0.847) and 0.734 (95% CI: 0.619-0.850), respectively. AUC of combined detection with the above indexes was 0.886 (95% CI: 0.809-0.964), greater than that of single index (P<0.05). Conclusion Chest CT combined with hs-CRP, D-D and PCT can effectively predict prognosis and guide the formulation of clinical protocols in patients with severe pneumonia.

Objective To investigate whether the extracellular volume fraction (ECV) determined using enhanced computed tomography (CT) can predict the pathologic grade of gastric cancer. Methods A total of 50 patients with gastric cancer who underwent surgery and were pathologically confirmed at Ningxia Medical University General Hospital from January 2023 to October 2024, and who had undergone enhanced CT scans before surgery, were analyzed. The plain, arterial, venous, and balance phase values were recorded, and the absolute contrast-enhanced CT differences ΔS1=HU_{arterial phase} -HU_{normal scan},ΔS2=HU_{venous phase}-HU_{normal scan},ΔS3=HU_{equilibrium phase}-HU_{normal scan} were obtained. The ECV of the primary lesion was calculated by measuring the CT values of the regions of interest in the plain and balance phases.Patients were allocated to either a low-grade or a high-grade group based on the histologic grading standard for gastric cancer(World Health Organization 2022 edition). The differences in the parameters between the two groups were evaluated for statistical significance. ROC curve was used to evaluate the diagnostic efficiency. Results The mean age of the 50 patients included [17 patients in the high-grade group (34.0%) and 33 patients in the low-grade group (66.0%)]had an average age of 62.54 years.There were significant differences between arterial stage (P=0.002), ΔS1 (P=0.002), ΔS3 (P=0.017), ECV (P<0.001) and pathological grade. The ROC curve demonstrated that the best efficacy in evaluating the pathologic grade of gastric cancer was achieved by ECV, with an area under the curve of 0.785 (95% CI: 0.646-0.924). The diagnostic threshold was 29.1%, sensitivity was 94.1%, and specificity was 36.4%. Conclusion The use of enhanced CT to obtain ECV is helpful in predicting the pathologic grade of gastric cancer.

Breast cancer is the malignant tumor with the highest incidence in women worldwide. Therefore, early detection and early diagnosis are the keys to prolong survival, but there are limitations in traditional imaging diagnostic methods. In recent years, artificial intelligence (AI) technology has significantly improved the accuracy and efficiency of breast cancer imaging diagnosis through deep learning and image processing. In breast ultrasound, mammography, Breast MRI and emerging imaging technologies, AI can promote the development of medical imaging through lesion detection, classification, image enhancement, risk prediction and clinical decision support. However, the clinical translation of AI still faces challenges such as data standardization, algorithm generalization, and ethical compliance. In the future, it is necessary to strengthen multi-center cooperation, promote technological innovation, and improve ethical regulations, to ensure that it can truly meet clinical needs, so as to promote the intelligence, precision and universality of breast cancer diagnosis and treatment. This paper provides a systematic review of the comparative advantages and limitations of AI versus conventional imaging methods in the diagnosis, treatment, and prognosis prediction of breast diseases. It explores feasible pathways for clinical translation and future development directions, while also offering insights into the prospective applications of AI in breast disease diagnosis and treatment.

Breast cancer ranks among the most prevalent malignancies in women globally, with consistently high incidence and mortality rates. Accurate assessment of axillary lymph node (ALN) status critically informs clinical decision-making and prognosis prediction. Although ALN dissection and sentinel lymph node biopsy remain the diagnostic gold standard, both procedures carry inherent limitations including surgical invasiveness and false-negative rates. Consequently, precise preoperative prediction of ALN status remains an unmet clinical need. Techniques such as mammography, CT, and breast MRI have advanced non-invasive evaluation. Radiomics and deep learning methodologies are now integral to ALN metastasis research in breast cancer. Notably, radiomics and deep learning models based on CT and MRI demonstrate robust performance in detecting ALN metastasis, achieving significant AUC values. This article systematically reviews recent advances in CT and MRI radiomics for predicting axillary lymph node metastasis in breast cancer.

Non-small cell lung cancer (NSCLC) is one of the leading causes of cancer-related deaths globally. Occult lymph node metastasis poses a significant challenge to the diagnosis, treatment planning, and prognosis of NSCLC patients. Early detection of occult lymph node metastasis through imaging can help select the optimal treatment method, thereby improving patient outcomes and ultimately contributing to reducing NSCLC-related mortality. This review discusses recent advancements in imaging features and radiomics for the early identification of occult lymph node metastasis in NSCLC, as well as the interpretability methods of radiomics.

Lung cancer, characterized by its complex and diverse etiology and subtle early clinical symptoms, has led to a consistently high incidence rate and a continuous increase year after year.CT imaging technology, due to its extremely high detection rate and ability to non-invasively and conveniently identify suspected lung lesions, is currently the best method for early lung cancer screening. This review will systematically elaborate on the advantages and disadvantages of two-dimensional reconstruction technology and three-dimensional reconstruction technology in CT images, and focus on the analysis of the clinical application value of three-dimensional reconstruction technology in the diagnosis and treatment of pulmonary nodules in chest CT images.The purpose is to select the most suitable reconstruction method for the diagnosis of clinical lung diseases, and to provide a reference for the innovation of three-dimensional reconstruction technology of lung nodules in chest CT in the future.

Breast cancer remains one of the leading causes of cancer-related mortality among women worldwide. Neoadjuvant chemotherapy (NAC) has become an essential component of contemporary breast cancer treatment. However, conventional methods for assessing NAC efficacy are often limited by subjectivity and suboptimal accuracy, underscoring the urgent need for more objective and reliable evaluation strategies. In recent years, AI, particularly radiomics and deep learning, has driven significant advances in medical imaging analysis. MRI-based radiomics combined with DL has demonstrated the ability to extract high-dimensional features from imaging data that are invisible to the human eye. These features can capture subtle microstructural alterations within tumors, characterize biological phenotypes, and quantify intratumoral heterogeneity, thereby substantially improving the precision of treatment response evaluation. This review highlights recent progress in the application of MRI-based AI technologies for predicting NAC response in breast cancer, aiming to facilitate the translation of these techniques from theory to clinical practice and to provide a scientific foundation for advancing precision and personalized oncology care.

Deep vein thrombosis (DVT) is a condition caused by abnormal blood clotting within deep veins. As a common peripheral vascular disease, its annual incidence rate is approximately 0.1%. DVT fragments can easily lead to pulmonary embolism. Additionally, approximately 20%-50% of patients develop post-thrombotic syndrome later on, severely impacting their quality of life. Therefore, early recognition and management of DVT are crucial for preventing complications such as pulmonary embolism. This review summarizes various imaging diagnostic techniques for DVT, including DSA, US, CT, and MRI, alongside deep learning-based medical image analysis methods for thrombus detection. The aim is to provide evidence-based guidance for DVT prevention and treatment.

With the extension of life expectancy, population aging and the increasing healthcare demands of elderly patients have become pressing realities. Aged patients often experience diminished physiological reserve and progressive deterioration across multiple systems including physiological, metabolic, and sensory functions resulting in reduced tolerance to intravenous anesthetics. Ciprofol is a novel non-barbiturate intravenous anesthetic independently developed and recently approved in China. Existing evidence indicates that ciprofol exhibits characteristics such as rapid onset, faster recovery, reduced injection pain, higher potency, and a wider safety window. However, most studies to date are small, single-center investigations lacking rigorous evidence grading, particularly in the context of clinical application in elderly patients. Based on current research, this review summarizes the clinical use of ciprofol in the elderly population, with the aim of supporting its rational administration in this demographic and proposing directions for future research.