Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.09.023
Yanghua Fan , Shuaiwei Guo , Chuming Tao , Hua Fang , Anna Mou , Ming Feng , Zhen Wu
Rationale and Objectives
The first-line treatment for prolactinoma is drug therapy with dopamine agonists (DAs). However, some patients with resistance to DA treatment should prioritize surgical treatment. Therefore, it is crucial to accurately identify the drug treatment response of prolactinoma before treatment. The present study was performed to determine the DA treatment response of prolactinoma using a clinical radiomic model that incorporated radiomic and clinical features before treatment.
Materials and Methods
In total, 255 patients diagnosed with prolactinoma were retrospectively divided to training and validation sets. An elastic net algorithm was used to screen the radiomic features, and a fusion radiomic model was established. A clinical radiomic model was then constructed to integrate the fusion radiomic model and the most important clinical features through multivariate logistic regression analysis for individual prediction. The calibration, discrimination, and clinical applicability of the established models were evaluated. 60 patients with prolactinoma from other centers were used to validate the performance of the constructed model.
Results
The fusion radiomic model was constructed from three significant radiomic features, and the area under the curve in the training set and validation set was 0.930 and 0.910, respectively. The clinical radiomic model was constructed using the radiomic model and three clinical features. The model exhibited good recognition and calibration abilities as evidenced by its area under the curve of 0.96, 0.92, and 0.92 in the training, validation, and external multicenter validation set, respectively. Analysis of the decision curve showed that the fusion radiomic model and clinical radiomic model had good clinical application value for DA treatment response prediction in patients with prolactinoma.
Conclusion
Our clinical radiomic model demonstrated high sensitivity and excellent performance in predicting DA treatment response in prolactinoma. This model holds promise for the noninvasive development of individualized diagnosis and treatment strategies for patients with prolactinoma.
{"title":"Noninvasive Radiomics Approach Predicts Dopamine Agonists Treatment Response in Patients with Prolactinoma: A Multicenter Study","authors":"Yanghua Fan , Shuaiwei Guo , Chuming Tao , Hua Fang , Anna Mou , Ming Feng , Zhen Wu","doi":"10.1016/j.acra.2024.09.023","DOIUrl":"10.1016/j.acra.2024.09.023","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>The first-line treatment for prolactinoma is drug therapy with dopamine agonists (DAs). However, some patients with resistance to DA treatment should prioritize surgical treatment. Therefore, it is crucial to accurately identify the drug treatment response of prolactinoma before treatment. The present study was performed to determine the DA treatment response of prolactinoma using a clinical radiomic model that incorporated radiomic and clinical features before treatment.</div></div><div><h3>Materials and Methods</h3><div>In total, 255 patients diagnosed with prolactinoma were retrospectively divided to training and validation sets. An elastic net algorithm was used to screen the radiomic features, and a fusion radiomic model was established. A clinical radiomic model was then constructed to integrate the fusion radiomic model and the most important clinical features through multivariate logistic regression analysis for individual prediction. The calibration, discrimination, and clinical applicability of the established models were evaluated. 60 patients with prolactinoma from other centers were used to validate the performance of the constructed model.</div></div><div><h3>Results</h3><div>The fusion radiomic model was constructed from three significant radiomic features, and the area under the curve in the training set and validation set was 0.930 and 0.910, respectively. The clinical radiomic model was constructed using the radiomic model and three clinical features. The model exhibited good recognition and calibration abilities as evidenced by its area under the curve of 0.96, 0.92, and 0.92 in the training, validation, and external multicenter validation set, respectively. Analysis of the decision curve showed that the fusion radiomic model and clinical radiomic model had good clinical application value for DA treatment response prediction in patients with prolactinoma.</div></div><div><h3>Conclusion</h3><div>Our clinical radiomic model demonstrated high sensitivity and excellent performance in predicting DA treatment response in prolactinoma. This model holds promise for the noninvasive development of individualized diagnosis and treatment strategies for patients with prolactinoma.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 612-623"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142331772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.09.049
Zexing He , Kaibin Fang , Xiaocong Lin , ChengHao Xiang , Yuanzhe Li , Nianlai Huang , XuJun Hu , Zekai Chen , Zhangsheng Dai
<div><h3>Rationale and Objectives</h3><div>Rotator cuff injury is a common ailment in the musculoskeletal system, with the subscapularis muscle being the largest and most robust muscle of the rotator cuff. The occurrence of subscapularis muscle tears is more frequent than previously reported. The main objective of this research is to harness the power of artificial intelligence to enhance the precision in diagnosing subscapularis muscle injuries via magnetic resonance imaging of the shoulder joint, prior to surgical intervention. This study seeks to integrate advanced artificial intelligence algorithms to analyze magnetic resonance imaging data, aiming to provide more accurate preoperative assessments, which can potentially lead to better surgical outcomes and patient care and promote technological progress in the field of medical imaging analysis.</div></div><div><h3>Method</h3><div>This is a multicenter study that involves 324 patients from a major medical center serving as both the training and testing groups, with an additional 60 patients from two other medical centers comprising the verifying group. The imaging protocol for all these subjects included a series of shoulder magnetic resonance imaging scans: T1-weighted coronal sequences, T2-weighted coronal, axial, and sagittal images. These comprehensive imaging modalities were utilized to thoroughly examine the shoulder joint's anatomical details and to detect any signs of subscapularis muscle damage. To enhance the diagnostic accuracy before surgical procedures, radiomic analysis was employed. This technique involves the extraction of a multitude of quantitative features from the magnetic resonance imaging, which can provide a more nuanced and data-driven approach to identifying subscapularis muscle injuries. The integration of radiomics in this study aims to offer a more precise preoperative assessment, potentially leading to improved surgical planning and patient outcomes.</div></div><div><h3>Result</h3><div>In the course of this study, a comprehensive extraction of 1197 radiomic features was performed for each imaging modality of every patient. The coronal T1-weighted modality, when assessed within the internal verifying cohort, delivered a diagnostic accuracy of 0.766, coupled with an AUC of 0.803. In the case of the T2-weighted modality, the coronal planes exhibited a diagnostic accuracy of 0.781 and an AUC of 0.844. The axial T2-weighted images recorded an accuracy of 0.719 and an AUC of 0.761, while the sagittal T2-weighted images scored an accuracy of 0.766 and an AUC of 0.821. The amalgamation of these imaging techniques through a multimodal strategy markedly enhanced the accuracy to 0.828, with an AUC of 0.916 for the internal verifying group. The diagnostic performance of the coronal T1-weighted modality in the external verifying cohort yielded an accuracy of 0.833, with an area under the curve (AUC) of 0.819. For the T2-weighted modality, the coronal imaging demonstrated an acc
{"title":"Enhancing Preoperative Diagnosis of Subscapular Muscle Injuries with Shoulder MRI-based Multimodal Radiomics","authors":"Zexing He , Kaibin Fang , Xiaocong Lin , ChengHao Xiang , Yuanzhe Li , Nianlai Huang , XuJun Hu , Zekai Chen , Zhangsheng Dai","doi":"10.1016/j.acra.2024.09.049","DOIUrl":"10.1016/j.acra.2024.09.049","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Rotator cuff injury is a common ailment in the musculoskeletal system, with the subscapularis muscle being the largest and most robust muscle of the rotator cuff. The occurrence of subscapularis muscle tears is more frequent than previously reported. The main objective of this research is to harness the power of artificial intelligence to enhance the precision in diagnosing subscapularis muscle injuries via magnetic resonance imaging of the shoulder joint, prior to surgical intervention. This study seeks to integrate advanced artificial intelligence algorithms to analyze magnetic resonance imaging data, aiming to provide more accurate preoperative assessments, which can potentially lead to better surgical outcomes and patient care and promote technological progress in the field of medical imaging analysis.</div></div><div><h3>Method</h3><div>This is a multicenter study that involves 324 patients from a major medical center serving as both the training and testing groups, with an additional 60 patients from two other medical centers comprising the verifying group. The imaging protocol for all these subjects included a series of shoulder magnetic resonance imaging scans: T1-weighted coronal sequences, T2-weighted coronal, axial, and sagittal images. These comprehensive imaging modalities were utilized to thoroughly examine the shoulder joint's anatomical details and to detect any signs of subscapularis muscle damage. To enhance the diagnostic accuracy before surgical procedures, radiomic analysis was employed. This technique involves the extraction of a multitude of quantitative features from the magnetic resonance imaging, which can provide a more nuanced and data-driven approach to identifying subscapularis muscle injuries. The integration of radiomics in this study aims to offer a more precise preoperative assessment, potentially leading to improved surgical planning and patient outcomes.</div></div><div><h3>Result</h3><div>In the course of this study, a comprehensive extraction of 1197 radiomic features was performed for each imaging modality of every patient. The coronal T1-weighted modality, when assessed within the internal verifying cohort, delivered a diagnostic accuracy of 0.766, coupled with an AUC of 0.803. In the case of the T2-weighted modality, the coronal planes exhibited a diagnostic accuracy of 0.781 and an AUC of 0.844. The axial T2-weighted images recorded an accuracy of 0.719 and an AUC of 0.761, while the sagittal T2-weighted images scored an accuracy of 0.766 and an AUC of 0.821. The amalgamation of these imaging techniques through a multimodal strategy markedly enhanced the accuracy to 0.828, with an AUC of 0.916 for the internal verifying group. The diagnostic performance of the coronal T1-weighted modality in the external verifying cohort yielded an accuracy of 0.833, with an area under the curve (AUC) of 0.819. For the T2-weighted modality, the coronal imaging demonstrated an acc","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 907-915"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142382296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.09.042
Adarsh Ghosh MD , Hailong Li PhD , Andrew T. Trout MD
Introduction
Original research in radiology often involves handling large datasets, data manipulation, statistical tests, and coding. Recent studies show that large language models (LLMs) can solve bioinformatics tasks, suggesting their potential in radiology research. This study evaluates an LLM's ability to provide statistical and deep learning solutions and code for radiology research.
Materials and Methods
We used web-based chat interfaces available for ChatGPT-4o, ChatGPT-3.5, and Google Gemini.
Experiment 1: Biostatistics and Data Visualization
We assessed each LLMs' ability to suggest biostatistical tests and generate R code for the same using a Cancer Imaging Archive dataset. Prompts were based on statistical analyses from a peer-reviewed manuscript. The generated code was tested in R Studio for correctness, runtime errors and the ability to generate the requested visualization.
Experiment 2: Deep Learning
We used the RSNA-STR Pneumonia Detection Challenge dataset to evaluate ChatGPT-4o and Gemini’s ability to generate Python code for transformer-based image classification models (Vision Transformer ViT-B/16). The generated code was tested in a Jupiter Notebook for functionality and run time errors.
Results
Out of the 8 statistical questions posed, correct statistical answers were suggested for 7 (ChatGPT-4o), 6 (ChatGPT-3.5), and 5 (Gemini) scenarios. The R code output by ChatGPT-4o had fewer runtime errors (6 out of the 7 total codes provided) compared to ChatGPT-3.5 (5/7) and Gemini (5/7). Both ChatGPT4o and Gemini were able to generate visualization requested with a few run time errors. Iteratively copying runtime errors from the code generated by ChatGPT4o into the chat helped resolve them. Gemini initially hallucinated during code generation but was able to provide accurate code on restarting the experiment.
ChatGPT4-o and Gemini successfully generated initial Python code for deep learning tasks. Errors encountered during implementation were resolved through iterations using the chat interface, demonstrating LLM utility in providing baseline code for further code refinement and resolving run time errors.
Conclusion
LLMs can assist in coding tasks for radiology research, providing initial code for data visualization, statistical tests, and deep learning models helping researchers with foundational biostatistical knowledge. While LLM can offer a useful starting point, they require users to refine and validate the code and caution is necessary due to potential errors, the risk of hallucinations and data privacy regulations.
Summary statement
LLMs can help with coding and statistical problems in radiology research. This can help primary authors trouble shoot coding needed in radiology research.
{"title":"Large Language Models can Help with Biostatistics and Coding Needed in Radiology Research","authors":"Adarsh Ghosh MD , Hailong Li PhD , Andrew T. Trout MD","doi":"10.1016/j.acra.2024.09.042","DOIUrl":"10.1016/j.acra.2024.09.042","url":null,"abstract":"<div><h3>Introduction</h3><div>Original research in radiology often involves handling large datasets, data manipulation, statistical tests, and coding. Recent studies show that large language models (LLMs) can solve bioinformatics tasks, suggesting their potential in radiology research. This study evaluates an LLM's ability to provide statistical and deep learning solutions and code for radiology research.</div></div><div><h3>Materials and Methods</h3><div>We used web-based chat interfaces available for ChatGPT-4o, ChatGPT-3.5, and Google Gemini.</div></div><div><h3>Experiment 1: Biostatistics and Data Visualization</h3><div>We assessed each LLMs' ability to suggest biostatistical tests and generate R code for the same using a Cancer Imaging Archive dataset. Prompts were based on statistical analyses from a peer-reviewed manuscript. The generated code was tested in R Studio for correctness, runtime errors and the ability to generate the requested visualization.</div></div><div><h3>Experiment 2: Deep Learning</h3><div>We used the RSNA-STR Pneumonia Detection Challenge dataset to evaluate ChatGPT-4o and Gemini’s ability to generate Python code for transformer-based image classification models (Vision Transformer ViT-B/16). The generated code was tested in a Jupiter Notebook for functionality and run time errors.</div></div><div><h3>Results</h3><div>Out of the 8 statistical questions posed, correct statistical answers were suggested for 7 (ChatGPT-4o), 6 (ChatGPT-3.5), and 5 (Gemini) scenarios. The R code output by ChatGPT-4o had fewer runtime errors (6 out of the 7 total codes provided) compared to ChatGPT-3.5 (5/7) and Gemini (5/7). Both ChatGPT4o and Gemini were able to generate visualization requested with a few run time errors. Iteratively copying runtime errors from the code generated by ChatGPT4o into the chat helped resolve them. Gemini initially hallucinated during code generation but was able to provide accurate code on restarting the experiment.</div><div>ChatGPT4-o and Gemini successfully generated initial Python code for deep learning tasks. Errors encountered during implementation were resolved through iterations using the chat interface, demonstrating LLM utility in providing baseline code for further code refinement and resolving run time errors.</div></div><div><h3>Conclusion</h3><div>LLMs can assist in coding tasks for radiology research, providing initial code for data visualization, statistical tests, and deep learning models helping researchers with foundational biostatistical knowledge. While LLM can offer a useful starting point, they require users to refine and validate the code and caution is necessary due to potential errors, the risk of hallucinations and data privacy regulations.</div></div><div><h3>Summary statement</h3><div>LLMs can help with coding and statistical problems in radiology research. This can help primary authors trouble shoot coding needed in radiology research.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 604-611"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142479918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.09.007
Sammar Ghannam MD, MPH , Varshaa Koneru MD , Patrick Karabon , Rachel Darling MD , Kenneth A. Kist MD , Pamela Otto MD , Thanh Van MD
Rationale and Objectives
The combination of functional biologic data and imaging appearance has the potential to add diagnostic information to help the radiologist evaluate breast masses in an efficient, effective, and cost-conscious manner. This is the first clinical evaluation of the Gen 2(Model 9100, 8101) Imagio® System to assess image quality with both the stand-alone internal ultrasound (IUS), ultrasound-only transducer, and the Optoacoustic/Ultrasound (OA/US) duplex probe 1, 2. This study assesses palpable and non-palpable breast abnormalities in patients who are referred for diagnostic breast ultrasound work-up. This study is intended to confirm the clinical acceptability of modifications made to the Imagio® System ultrasound component following Premarket Approval (PMA) of the Imagio® Gen 1 version.
Materials and Methods
This prospective, single-arm, non-randomized study included 38 patients presenting with a palpable lump and/or imaging abnormality detected at a single investigational site. Each patient had the breast, and if indicated, the axillary lymph nodes imaged with the Gen 2 Imagio® system.
Results
For patients with SenoGram®-predicted Likelihood of Malignancy (LOM) and pathology available (N = 23), observed sensitivity was 100.0% (9/9) with a confidence interval of (66.4%, 100.0%), using a SenoGram®-predicted False Negative Rate (FNR) cut-off of ≤ 2%. Observed specificity was 64.3% (9/14) (Confidence Interval: 35.1%, 87.2%), using a SenoGram®-predicted FNR cut-off of ≤ 2%. At 98% fixed sensitivity, the specificity (fSp) for OA/US + SG was 100.0% while it was 0.0% for IUS. The absolute gain in fSp was 100.0%.
Conclusion
Combining structure with morphology can increase specificity without decreasing sensitivity in a real-world setting.
{"title":"Exploring the Utility of Optoacoustic Imaging in Differentiation of Benign and Malignant Breast Masses: Gen 2 Study","authors":"Sammar Ghannam MD, MPH , Varshaa Koneru MD , Patrick Karabon , Rachel Darling MD , Kenneth A. Kist MD , Pamela Otto MD , Thanh Van MD","doi":"10.1016/j.acra.2024.09.007","DOIUrl":"10.1016/j.acra.2024.09.007","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>The combination of functional biologic data and imaging appearance has the potential to add diagnostic information to help the radiologist evaluate breast masses in an efficient, effective, and cost-conscious manner. This is the first clinical evaluation of the Gen 2(Model 9100, 8101) Imagio® System to assess image quality with both the stand-alone internal ultrasound (IUS), ultrasound-only transducer, and the Optoacoustic/Ultrasound (OA/US) duplex probe <span><span>1</span></span>, <span><span>2</span></span>. This study assesses palpable and non-palpable breast abnormalities in patients who are referred for diagnostic breast ultrasound work-up. This study is intended to confirm the clinical acceptability of modifications made to the Imagio® System ultrasound component following Premarket Approval (PMA) of the Imagio® Gen 1 version.</div></div><div><h3>Materials and Methods</h3><div>This prospective, single-arm, non-randomized study included 38 patients presenting with a palpable lump and/or imaging abnormality detected at a single investigational site. Each patient had the breast, and if indicated, the axillary lymph nodes imaged with the Gen 2 Imagio® system.</div></div><div><h3>Results</h3><div>For patients with SenoGram®-predicted Likelihood of Malignancy (LOM) and pathology available (N = 23), observed sensitivity was 100.0% (9/9) with a confidence interval of (66.4%, 100.0%), using a SenoGram®-predicted False Negative Rate (FNR) cut-off of ≤ 2%. Observed specificity was 64.3% (9/14) (Confidence Interval: 35.1%, 87.2%), using a SenoGram®-predicted FNR cut-off of ≤ 2%. At 98% fixed sensitivity, the specificity (fSp) for OA/US + SG was 100.0% while it was 0.0% for IUS. The absolute gain in fSp was 100.0%.</div></div><div><h3>Conclusion</h3><div>Combining structure with morphology can increase specificity without decreasing sensitivity in a real-world setting.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 634-650"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142480005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.10.006
Enes Gurun MD , Ahmet Veli Sanibas MD , Mertcan Tekgoz MD , Dilara Erdogan MD
Rationale and Objectives
We aimed to evaluate possible elasticity changes in the menisci of patients with type 2 diabetes mellitus using shear wave elastography (SWE).
Materials and Methods
The medial and lateral menisci of the right and left knee of 40 patients (20 males, 20 females) with type 2 diabetes mellitus and 40 healthy controls (20 males, 20 females) were evaluated between June 2024 and September 2024. All patients and the control group were evaluated with MRI for meniscal pathology. Medial and lateral meniscal thicknesses were measured in the coronal plane in grayscale US mode. In both groups, the SWE measurement range was set to 0–8.2 m/s and 0–200 kPa and 2 mm ROIs were placed on the medial and lateral meniscal bodies of both knees in the coronal plane. The stiffness values of the meniscus were measured three times and the mean value of these three measurements was recorded.
Results
There was no significant difference between meniscal thickness in diabetic patients and control group (p > 0.05). Bilateral meniscal stiffness values of diabetic patients were higher than the control group and there was a statistically significant difference (p < 0.05). There were moderate to strong positive correlations between meniscal stiffness values and fasting blood glucose and HA1c values in the diabetic patients(p < 0.05).
Conclusion
This is the first study to demonstrate that meniscus stiffness increases in diabetic patients. SWE is a quantitative imaging method that can be used to detect meniscal pathologies that may develop due to diabetes.
{"title":"Evaluation of Meniscus Elasticity with Shear Wave Elastography in Patients with Type 2 Diabetes Mellitus","authors":"Enes Gurun MD , Ahmet Veli Sanibas MD , Mertcan Tekgoz MD , Dilara Erdogan MD","doi":"10.1016/j.acra.2024.10.006","DOIUrl":"10.1016/j.acra.2024.10.006","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>We aimed to evaluate possible elasticity changes in the menisci of patients with type 2 diabetes mellitus using shear wave elastography (SWE).</div></div><div><h3>Materials and Methods</h3><div>The medial and lateral menisci of the right and left knee of 40 patients (20 males, 20 females) with type 2 diabetes mellitus and 40 healthy controls (20 males, 20 females) were evaluated between June 2024 and September 2024. All patients and the control group were evaluated with MRI for meniscal pathology. Medial and lateral meniscal thicknesses were measured in the coronal plane in grayscale US mode. In both groups, the SWE measurement range was set to 0–8.2<!--> <!-->m/s and 0–200<!--> <!-->kPa and 2 mm ROIs were placed on the medial and lateral meniscal bodies of both knees in the coronal plane. The stiffness values of the meniscus were measured three times and the mean value of these three measurements was recorded.</div></div><div><h3>Results</h3><div>There was no significant difference between meniscal thickness in diabetic patients and control group (p > 0.05). Bilateral meniscal stiffness values of diabetic patients were higher than the control group and there was a statistically significant difference (p < 0.05). There were moderate to strong positive correlations between meniscal stiffness values and fasting blood glucose and HA1c values in the diabetic patients(p < 0.05).</div></div><div><h3>Conclusion</h3><div>This is the first study to demonstrate that meniscus stiffness increases in diabetic patients. SWE is a quantitative imaging method that can be used to detect meniscal pathologies that may develop due to diabetes.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 916-921"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142569851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To establish a multimodal deep learning nomogram for predicting clinically significant prostate cancer in patients with gray-zone PSA levels.
Methods
This retrospective study enrolled 303 patients with pathological results between January 2018 and December 2022. Clinical variables and the PI-RADS v2.1 score were used to construct a clinical model. Radiomics and deep learning features from bp-MRI were used to develop a radiomics model with SVM and a deep learning model, respectively. A hybrid fusion approach was used to integrate the multimodal data and construct combined models (Comb.Rad.model and Comb.DL.model). The robustness of the radiomics model with XGBoost was validated and compared. Model efficacy was assessed through ROC curve and decision curve analysis. A nomogram was developed based on the best-performing model.
Results
The clinical model had AUCs of 0.845 and 0.779 in the training and testing set. The radiomics model with SVM and the deep learning model achieved AUCs of 0.825 and 0.933 in the training set and 0.811 and 0.907 in the testing set, respectively. The diagnostic performance of the combined models was significantly improved, with Comb.DL.model having a higher AUC than Comb.Rad.model in both the training (0.986 vs. 0.924, P = 0.008) and testing (0.965 vs. 0.859, P = 0.005) set. The diagnostic efficiency of both the radiomics model and Comb.Rad.model with XGBoost were comparable to that of SVM, confirming the robustness of the established model.
Conclusion
The integrated nomogram combining deep learning features, PI-RADS score, and clinical variables significantly outperformed the traditional radiomics and clinical models.
{"title":"A Multimodal Deep Learning Nomogram for the Identification of Clinically Significant Prostate Cancer in Patients with Gray-Zone PSA Levels: Comparison with Clinical and Radiomics Models","authors":"Tong Chen , Wei Hu , Yueyue Zhang , Chaogang Wei , Wenlu Zhao , Xiaohong Shen , Caiyuan Zhang , Junkang Shen","doi":"10.1016/j.acra.2024.10.009","DOIUrl":"10.1016/j.acra.2024.10.009","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To establish a multimodal deep learning nomogram for predicting clinically significant prostate cancer in patients with gray-zone PSA levels.</div></div><div><h3>Methods</h3><div>This retrospective study enrolled 303 patients with pathological results between January 2018 and December 2022. Clinical variables and the PI-RADS v2.1 score were used to construct a clinical model. Radiomics and deep learning features from bp-MRI were used to develop a radiomics model with SVM and a deep learning model, respectively. A hybrid fusion approach was used to integrate the multimodal data and construct combined models (Comb.Rad.model and Comb.DL.model). The robustness of the radiomics model with XGBoost was validated and compared. Model efficacy was assessed through ROC curve and decision curve analysis. A nomogram was developed based on the best-performing model.</div></div><div><h3>Results</h3><div>The clinical model had AUCs of 0.845 and 0.779 in the training and testing set. The radiomics model with SVM and the deep learning model achieved AUCs of 0.825 and 0.933 in the training set and 0.811 and 0.907 in the testing set, respectively. The diagnostic performance of the combined models was significantly improved, with Comb.DL.model having a higher AUC than Comb.Rad.model in both the training (0.986 vs. 0.924, P = 0.008) and testing (0.965 vs. 0.859, P = 0.005) set. The diagnostic efficiency of both the radiomics model and Comb.Rad.model with XGBoost were comparable to that of SVM, confirming the robustness of the established model.</div></div><div><h3>Conclusion</h3><div>The integrated nomogram combining deep learning features, PI-RADS score, and clinical variables significantly outperformed the traditional radiomics and clinical models.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 864-876"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142577170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.10.005
Sam Afshari , Jacob Lythgoe , Megan Zhou , Connor Barton , Andrew Warfield , Ryan Walsh , Abigail Hielscher Ph.D.
Rationale and Objectives
Competency in imaging is essential for physicians to diagnose and manage disease. Previously, the authors introduced radiology education in the anatomy lab. The present study transitioned the radiology education to the classroom with the primary goal of increasing engagement and clinical relevance.
Materials and Methods
To accomplish these objectives, a team of senior medical students, residents, a diagnostic radiologist, and an anatomist collaborated to design pre-work e-modules and active learning workshops focused on imaging five body regions. For three regions, interactive e-modules with built-in quizzes and videos were designed. PowerPoints were used for the other two regions. Pacsbin, a web-based Digital Imaging and Communications in Medicine viewer, was used as a platform to introduce students to the basics of windowing, scrolling and labeling images. Workshops focused on 3–4 cases which instructed groups of students to scroll through and label anatomical structures on scans uploaded to Pacsbin. A questionnaire seeking students’ feedback on the curriculum was given at the end of the course.
Results
Students indicated high satisfaction with the imaging curriculum, believing that it supported their anatomical knowledge. The majority of students preferred the e-modules as opposed to PowerPoints for learning the imaging anatomy. Pacsbin was most often used only during workshops. Students’ responses regarding their confidence with use Pacsbin were almost evenly distributed on a 4-point Likert scale.
Conclusions
Overall, this work presents an interdisciplinary way by which imaging can be incorporated into the pre-clinical medical curriculum in an engaging and clinically relevant manner.
{"title":"An Interdisciplinary Approach Toward Developing an Engaging and Clinically Relevant Medical Imaging Curriculum","authors":"Sam Afshari , Jacob Lythgoe , Megan Zhou , Connor Barton , Andrew Warfield , Ryan Walsh , Abigail Hielscher Ph.D.","doi":"10.1016/j.acra.2024.10.005","DOIUrl":"10.1016/j.acra.2024.10.005","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Competency in imaging is essential for physicians to diagnose and manage disease. Previously, the authors introduced radiology education in the anatomy lab. The present study transitioned the radiology education to the classroom with the primary goal of increasing engagement and clinical relevance.</div></div><div><h3>Materials and Methods</h3><div>To accomplish these objectives, a team of senior medical students, residents, a diagnostic radiologist, and an anatomist collaborated to design pre-work e-modules and active learning workshops focused on imaging five body regions. For three regions, interactive e-modules with built-in quizzes and videos were designed. PowerPoints were used for the other two regions. Pacsbin, a web-based Digital Imaging and Communications in Medicine viewer, was used as a platform to introduce students to the basics of windowing, scrolling and labeling images. Workshops focused on 3–4 cases which instructed groups of students to scroll through and label anatomical structures on scans uploaded to Pacsbin. A questionnaire seeking students’ feedback on the curriculum was given at the end of the course.</div></div><div><h3>Results</h3><div>Students indicated high satisfaction with the imaging curriculum, believing that it supported their anatomical knowledge. The majority of students preferred the e-modules as opposed to PowerPoints for learning the imaging anatomy. Pacsbin was most often used only during workshops. Students’ responses regarding their confidence with use Pacsbin were almost evenly distributed on a 4-point Likert scale.</div></div><div><h3>Conclusions</h3><div>Overall, this work presents an interdisciplinary way by which imaging can be incorporated into the pre-clinical medical curriculum in an engaging and clinically relevant manner.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 1127-1137"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142592053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aim of this study was to analyze demographic data of academic radiology faculty to assess rank equity by gender and race/ethnicity and trends from 2000 to 2023.
Methods
Data was collected from the AAMC Faculty Salary Roster, which collects information for self-reported gender and race/ethnicity for faculty at different ranks in U.S. medical schools. To determine parity between faculty ranks across gender and race/ethnicity, rank equity index (REI) was calculated for associate/assistant, professor/associate, and professor/assistant professor comparisons.
Results
The percentage of women faculty increased from 23.6% in 2000 to 30% in 2023. REI increased steadily for women, and White women reached parity in 2023 for Associate/Assistant comparison but not for Professor/Assistant. REI remained low for Asian and URM women (0.67–0.69 for Professor/Assistant comparison). Only Asian men reached parity for Professor/Assistant comparison, and REI decreased for URM men over the study period. Black faculty had a modest improvement in REI from 2000 (0.41) to 2009 (0.67) but remained unchanged since then (0.67 in 2023).
Conclusion
Advancement along the academic ladder has been uneven in academic radiology. While rank equity for women has improved over time, for URM and Asian women it remains substantially below parity. URM men have actually seen a decline in rank equity across ranks. Further efforts are needed to identify barriers to recruitment, retention, and promotion for these sub-groups in academic radiology and create interventions that diversify radiology faculty at all ranks.
理由和目标：本研究旨在分析放射学学术教师的人口统计学数据，以评估按性别和种族/族裔划分的职级公平性以及 2000 年至 2023 年的趋势：数据收集自美国医学会教职员工薪资名册，该名册收集了美国医学院不同级别教职员工自我报告的性别和种族/族裔信息。为了确定不同性别和种族/族裔教员职级之间的均等性，计算了副教授/助理教授、教授/副教授和教授/助理教授的职级公平指数（REI）：女教师的比例从 2000 年的 23.6% 增加到 2023 年的 30%。女性的 REI 稳步上升，白人女性在 2023 年的副教授/助理比较中达到了均等，但在教授/助理比较中没有达到。亚裔和统招女生的 REI 仍然较低（教授/助理比较为 0.67-0.69）。只有亚裔男性在教授/助理比较中达到了均等，而在研究期间，统 一种族男性的 REI 有所下降。黑人教师的 REI 从 2000 年（0.41）到 2009 年（0.67）略有提高，但此后保持不变（2023 年为 0.67）：结论：放射学学术梯队的发展并不平衡。尽管随着时间的推移，女性的职级公平性有所改善，但对于统招研究生和亚裔女性而言，其职级公平性仍然远远低于均等水平。实际上，统招男性的职级公平性有所下降。我们需要进一步努力，找出放射学学术界这些亚群体在招聘、留用和晋升方面的障碍，并制定干预措施，使各级放射学教职员工多样化。
{"title":"Trends in Faculty Advancement for Underrepresented groups in Academic Radiology","authors":"Ajay Malhotra MD, MMM , Dheeman Futela MBBS , Shadi Ebrahimian MD , Siddhi Singhania , Seyedmehdi Payabvash MD , John E. Jordan MD, MPP, FACR , Dheeraj Gandhi MD, FACR","doi":"10.1016/j.acra.2024.10.030","DOIUrl":"10.1016/j.acra.2024.10.030","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>The aim of this study was to analyze demographic data of academic radiology faculty to assess rank equity by gender and race/ethnicity and trends from 2000 to 2023.</div></div><div><h3>Methods</h3><div>Data was collected from the AAMC Faculty Salary Roster, which collects information for self-reported gender and race/ethnicity for faculty at different ranks in U.S. medical schools. To determine parity between faculty ranks across gender and race/ethnicity, rank equity index (REI) was calculated for associate/assistant, professor/associate, and professor/assistant professor comparisons.</div></div><div><h3>Results</h3><div>The percentage of women faculty increased from 23.6% in 2000 to 30% in 2023. REI increased steadily for women, and White women reached parity in 2023 for Associate/Assistant comparison but not for Professor/Assistant. REI remained low for Asian and URM women (0.67–0.69 for Professor/Assistant comparison). Only Asian men reached parity for Professor/Assistant comparison, and REI decreased for URM men over the study period. Black faculty had a modest improvement in REI from 2000 (0.41) to 2009 (0.67) but remained unchanged since then (0.67 in 2023).</div></div><div><h3>Conclusion</h3><div>Advancement along the academic ladder has been uneven in academic radiology. While rank equity for women has improved over time, for URM and Asian women it remains substantially below parity. URM men have actually seen a decline in rank equity across ranks. Further efforts are needed to identify barriers to recruitment, retention, and promotion for these sub-groups in academic radiology and create interventions that diversify radiology faculty at all ranks.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Pages 722-727"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142631740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2023.10.058
Longsheng Wang MA , Liwei Zou MA , Qi Chen BA , Lianzi Su MA , Jiajia Xu MA , Ru Zhao MA , Yanqi Shan MA , Qing Zhang MA , Zhimin Zhai MD , Xijun Gong MA , Hong Zhao MD , Fangbiao Tao MD , Suisheng Zheng BA
{"title":"Corrigendum to “Gray Matter Structural Network Disruptions in Survivors of Acute Lymphoblastic Leukemia with Chemotherapy Treatment”","authors":"Longsheng Wang MA , Liwei Zou MA , Qi Chen BA , Lianzi Su MA , Jiajia Xu MA , Ru Zhao MA , Yanqi Shan MA , Qing Zhang MA , Zhimin Zhai MD , Xijun Gong MA , Hong Zhao MD , Fangbiao Tao MD , Suisheng Zheng BA","doi":"10.1016/j.acra.2023.10.058","DOIUrl":"10.1016/j.acra.2023.10.058","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Page 1160"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138177827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.acra.2024.10.008
Grace G. Zhu MD , Alexander Y. Xie FSA , Fatima Elahi DO , Cameron Overfield MD , Jordan Mackner BS , Amit Chakraborty MD , Richard H. Wiggins MD
{"title":"Corrigendum to ‘RadDiscord’s Big Bang: Perspectives and Impact of Creation of a Successful Radiology Education Community’ Academic Radiology/ Volume 31, Issue 2, February 2024/ pages 390-398","authors":"Grace G. Zhu MD , Alexander Y. Xie FSA , Fatima Elahi DO , Cameron Overfield MD , Jordan Mackner BS , Amit Chakraborty MD , Richard H. Wiggins MD","doi":"10.1016/j.acra.2024.10.008","DOIUrl":"10.1016/j.acra.2024.10.008","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 2","pages":"Page 1164"},"PeriodicalIF":3.8,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142480000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号