Indian Journal of Radiology and Imaging最新文献

{"title":"Enhancing Radiological Reporting in Head and Neck Cancer: Converting Free-Text CT Scan Reports to Structured Reports Using Large Language Models.","authors":"Amit Gupta, Hema Malhotra, Amit K Garg, Krithika Rangarajan","doi":"10.1055/s-0044-1788589","DOIUrl":"10.1055/s-0044-1788589","url":null,"abstract":"<p><p><b>Objective</b>  The aim of this study was to assess efficacy of large language models (LLMs) for converting free-text computed tomography (CT) scan reports of head and neck cancer (HNCa) patients into a structured format using a predefined template. <b>Materials and Methods</b>  A retrospective study was conducted using 150 CT reports of HNCa patients. A comprehensive structured reporting template for HNCa CT scans was developed, and the Generative Pre-trained Transformer 4 (GPT-4) was initially used to convert 50 CT reports into a structured format using this template. The generated structured reports were then evaluated by a radiologist for instances of missing or misinterpreted information and any erroneous additional details added by GPT-4. Following this assessment, the template was refined for improved accuracy. This revised template was then used for conversion of 100 other HNCa CT reports into structured format using GPT-4. These reports were then reevaluated in the same manner. <b>Results</b>  Initially, GPT-4 successfully converted all 50 free-text reports into structured reports. However, there were 10 places with missing information: tracheostomy tube ( <i>n</i>  = 3), noninclusion of involvement of sternocleidomastoid muscle ( <i>n</i>  = 2), extranodal tumor extension ( <i>n</i>  = 3), and contiguous involvement of the neck structures by nodal mass rather than the primary ( <i>n</i>  = 2). Few instances of nonsuspicious lung nodules were misinterpreted as metastases ( <i>n</i>  = 2). GPT-4 did not indicate any erroneous additional findings. Using the revised reporting template, GPT-4 converted all the 100 CT reports into a structured format with no repeated or additional mistakes. <b>Conclusion</b>  LLMs can be used for structuring free-text radiology reports using plain language prompts and a simple yet comprehensive reporting template. <b>Key Points</b> Structured radiology reports in oncological patients, although advantageous, are not used widely in practice due to perceived drawbacks like interference with routine radiology workflow and scan interpretation.We found that GPT-4 is highly efficient in converting conventional CT reports of HNCa patients to structured reports using a predefined template.This application of LLMs in radiology can help in enhancing the acceptability and clinical utility of structured radiology reports in oncological imaging. <b>Summary Statement</b> Large language models can successfully and accurately convert conventional radiology reports for oncology scans into a structured format using a comprehensive predefined template and thus can enhance the utility and integration of these reports in routine clinical practice.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"35 1","pages":"43-49"},"PeriodicalIF":0.9,"publicationDate":"2024-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11651842/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142856139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Delayed Presentation of Baffle Obstruction in an Adult Post-Mustard Repair of Transposition: Computed Tomography Demonstration.","authors":"Debanjan Nandi, Amarinder Singh Malhi, Manish Shaw, Sanjeev Kumar","doi":"10.1055/s-0044-1787685","DOIUrl":"https://doi.org/10.1055/s-0044-1787685","url":null,"abstract":"<p><p>Atrial switch surgery is performed in patients with transposition of the great arteries. One of the complications of this surgery is obstruction of the baffle created. We describe the computed tomography findings of one such case where there was delayed presentation of recurrent Mustard baffle obstruction in addition to pulmonary venous drainage obstruction in an adult previously operated on for intra-atrial repair of transposition of the great arteries.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 4","pages":"757-760"},"PeriodicalIF":0.9,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11419738/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142332164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the Veil: Deflux Injection Masquerading as Calculus at the Vesicoureteric Junction.","authors":"Josh K Dasan, Antony Augustine, J Chandra Singh, Anuradha Chandramohan","doi":"10.1055/s-0044-1787779","DOIUrl":"https://doi.org/10.1055/s-0044-1787779","url":null,"abstract":"","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 4","pages":"795-798"},"PeriodicalIF":0.9,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11419740/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142332076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Total Anomalous Pulmonary Venous Connection with Rare Direct Right Atrial Drainage and Unprecedented Array of Coexistent Multisystem Variations.","authors":"Arun Sharma, Dollphy Garg, Shivali Arya, Sanjeev Hanumantacharya Naganur, Manphool Singhal","doi":"10.1055/s-0044-1787684","DOIUrl":"https://doi.org/10.1055/s-0044-1787684","url":null,"abstract":"<p><p>Total anomalous pulmonary venous connection (TAPVC) is anomalous drainage of all pulmonary veins into systemic circulation. The intracardiac type typically entails the drainage of all the pulmonary veins into the right atrium, via the coronary sinus. The connection of the pulmonary veins directly into the right atrium is exceptionally rare and has been primarily reported with right atrial isomerism. Herein, we presented a remarkable case of TAPVC in a 10-year-old male child, distinguished by an unconventional drainage of all the pulmonary veins directly into the right atrium, with normal coronary sinus and absent right atrial isomerism. Intriguingly, computed tomography imaging revealed a combination of incredibly rare coexistent pulmonary, vascular, and skeletal anomalies. These anomalies included absence of pulmonary fissures in the right lung, presence of left circumflex aortic arch with bovine branching pattern, bilateral cervical ribs, and C7 vertebral fusion anomalies. To our knowledge, this unique combination of coexistent anomalies has not been previously reported in scientific literature in the background of rare drainage pattern of TAPVC.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 4","pages":"761-764"},"PeriodicalIF":0.9,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11419768/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142332075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CSF Flow Dynamics-Is There a Need for Normative Imaging Data?","authors":"Jitender Saini, Shikha Awasthi","doi":"10.1055/s-0044-1781468","DOIUrl":"10.1055/s-0044-1781468","url":null,"abstract":"","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 2","pages":"199"},"PeriodicalIF":0.9,"publicationDate":"2024-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10972647/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140319872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Image-Guided Percutaneous Sclerotherapy for Orbital Low-Flow Malformation: Our Experience.","authors":"Gautam Lokdarshi, Shwetam Kumar, Subhash Kumar, Nitu Nitu, Nripen Gaur","doi":"10.1055/s-0044-1779689","DOIUrl":"10.1055/s-0044-1779689","url":null,"abstract":"<p><p>For a safe sclerotherapy session to be completed in the orbital low-flow malformation (namely lymphovenous malformation or venolymphatic malformation), accurate identification of the target lesion for the drug injection is crucial. Regarding the dependability and viability of the injection approach, the authors have discussed their experiences with image-guided percutaneous sclerotherapy on a few patients.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 3","pages":"539-544"},"PeriodicalIF":0.9,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11188674/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141443740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Approach to Nonmass Lesions on Breast Ultrasound.","authors":"Supraja Laguduva Mohan, Ekta Dhamija, Richa Gauba","doi":"10.1055/s-0044-1779589","DOIUrl":"https://doi.org/10.1055/s-0044-1779589","url":null,"abstract":"<p><p>Nonmass lesions in breast ultrasound (US) are areas of altered echogenicity without definite margins or mass effect. However, these lesions may show calcifications, associated architectural distortion, or shadowing just like masses. They vary in their echogenicity, distribution, ductal or nonductal appearance and the associated features that can be seen in variety of benign and malignant pathologies. With no uniform definition or classification system, there is no standardized approach in further risk categorization and management strategies of these lesions. Malignant nonmass lesions are not uncommon and few sonographic features can help in differentiating benign and malignant pathologies. US-guided tissue sampling or lesion localization can be preferred in the nonmass lesions identified on second look US after magnetic resonance imaging or mammography. This article aims to describe various imaging patterns and attempts to provide an algorithmic approach to nonmass findings on breast US.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 4","pages":"677-687"},"PeriodicalIF":0.9,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11419763/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142332163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pretherapeutic PSMA PET-Derived Semiquantitative Parameters as Predictors of PSA Response in Patients with mCRPC Receiving [ ¹⁷⁷ Lu]Lu-PSMA-617 Radioligand Therapy.","authors":"Dheeratama Siripongsatian, Attapon Jantarato, Chetsadaporn Promteangtrong, Anchisa Kunawudhi, Peerapon Kiatkittikul, Natphimol Boonkawin, Sukanya Yaset, Sirinsuda Somboon, Chanisa Chotipanich","doi":"10.1055/s-0044-1779634","DOIUrl":"https://doi.org/10.1055/s-0044-1779634","url":null,"abstract":"<p><p><b>Objective</b>  [ ¹⁷⁷ Lu]Lu-prostate-specific membrane antigen (PSMA)-617 radioligand therapy (RLT) shows promise for metastatic castration-resistant prostate cancer (mCRPC) patients with positive PSMA positron emission tomography (PET) imaging. Identifying high-risk patients is crucial. We evaluated pretherapeutic PSMA PET-derived parameters to predict prostate-specific antigen (PSA) response in patients undergoing [ ¹⁷⁷ Lu]Lu-PSMA-617 RLT. <b>Materials and Methods</b>  We conducted a retrospective analysis among 27 patients (mean age: 71.0 ± 9.5 years; range: 52-85 years) who underwent PSMA PET/computed tomography (CT) and subsequent [ ¹⁷⁷ Lu]Lu-PSMA-617 RLT between March 2019 and January 2023. After excluding patients with liver metastases, the number of patients left for analysis was 21 (14 responders and 7 nonresponders). Tumors were semiautomatically delineated with calculation of total tumor volume (PSMA-TV), lesion uptake (PSMA-TLU = PSMA-TV * standardized uptake value [SUV]mean), and lesion quotient (PSMA-TLQ = PSMA-TV/SUVmean) for each patient. Semiquantitative parameters were analyzed only in patients with mCRPC and no liver metastasis. <b>Results</b>  In total, 17/27 patients (62.96%) had a decline in PSA levels; 15/27 patients (55.56%) experienced a decline of > 50%. Pretherapeutic PSMA PET/CT results revealed significant differences in PSMA-TV ( <i>p</i>  = 0.003), PSMA-TLU ( <i>p</i>  = 0.013), and PSMA-TLQ ( <i>p</i>  = 0.011) between responders and nonresponders. SUVmax was significantly correlated to the best percentage change in PSA response after ¹⁷⁷ Lu-PSMA-617 treatment ( <i>r</i>  = -0.79, <i>p</i>  = 0.006). No association was observed between PSMA-TV ( <i>p</i>  = 0.367), PSMA-TLU ( <i>p</i>  = 0.128), and PSMA-TLQ ( <i>p</i>  = 0.556), with the best percentage change in PSA response after ¹⁷⁷ Lu-PSMA-617 therapy. <b>Conclusion</b>  Pretherapeutic PSMA PET-derived PSMA-TV, PSMA-TLU, and PSMA-TLQ were significant negative predictors of PSA response in patients with mCRPC and no liver metastasis receiving [ ¹⁷⁷ Lu]Lu-PSMA-617 RLT.</p>","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"34 4","pages":"579-587"},"PeriodicalIF":0.9,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11419753/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142332165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the Capability of ChatGPT, Google Bard, and Microsoft Bing in Solving Radiology Case Vignettes","authors":"Pradosh Kumar Sarangi, Ravi Kant Narayan, S. Mohakud, Aditi Vats, Debabrata Sahani, Himel Mondal","doi":"10.1055/s-0043-1777746","DOIUrl":"https://doi.org/10.1055/s-0043-1777746","url":null,"abstract":"Abstract Background  The field of radiology relies on accurate interpretation of medical images for effective diagnosis and patient care. Recent advancements in artificial intelligence (AI) and natural language processing have sparked interest in exploring the potential of AI models in assisting radiologists. However, limited research has been conducted to assess the performance of AI models in radiology case interpretation, particularly in comparison to human experts. Objective  This study aimed to evaluate the performance of ChatGPT, Google Bard, and Bing in solving radiology case vignettes (Fellowship of the Royal College of Radiologists 2A [FRCR2A] examination style questions) by comparing their responses to those provided by two radiology residents. Methods  A total of 120 multiple-choice questions based on radiology case vignettes were formulated according to the pattern of FRCR2A examination. The questions were presented to ChatGPT, Google Bard, and Bing. Two residents wrote the examination with the same questions in 3 hours. The responses generated by the AI models were collected and compared to the answer keys and explanation of the answers was rated by the two radiologists. A cutoff of 60% was set as the passing score. Results  The two residents (63.33 and 57.5%) outperformed the three AI models: Bard (44.17%), Bing (53.33%), and ChatGPT (45%), but only one resident passed the examination. The response patterns among the five respondents were significantly different ( p  = 0.0117). In addition, the agreement among the generative AI models was significant (intraclass correlation coefficient [ICC] = 0.628), but there was no agreement between the residents (Kappa = –0.376). The explanation of generative AI models in support of answer was 44.72% accurate. Conclusion  Humans exhibited superior accuracy compared to the AI models, showcasing a stronger comprehension of the subject matter. All three AI models included in the study could not achieve the minimum percentage needed to pass an FRCR2A examination. However, generative AI models showed significant agreement in their answers where the residents exhibited low agreement, highlighting a lack of consistency in their responses.","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":" 33","pages":""},"PeriodicalIF":0.6,"publicationDate":"2023-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139143939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bedside Ultrasound-Guided Percutaneous Cholecystostomy in Critically Ill Patients—Outcomes in 51 Patients","authors":"Rozil Gandhi, K. Gala, Mohd Shariq, Aditi Gandhi, Manish Gandhi, Amit Shah","doi":"10.1055/s-0043-1777744","DOIUrl":"https://doi.org/10.1055/s-0043-1777744","url":null,"abstract":"Abstract Purpose  The aim of this study was to report technical and clinical success of bedside ultrasound-guided percutaneous cholecystostomy (PC) tube placement in intensive care unit (ICU). Materials and Methods  This is a retrospective study of 51 patients (36 males:15 females, mean age: 67 years) who underwent ultrasound-guided PC from May 2015 to January 2020. The indication for cholecystostomy tube placement, comorbidities, imaging finding, technical success, clinical success, timing of surgery post-cholecystostomy tube placement, indwelling catheter time, complications, and follow-up were recorded. Results  Indications for cholecystostomy tube placement were acute calculous cholecystitis ( n  = 43; 84.3%), perforated cholecystitis ( n  = 5; 9.8%), and emphysematous cholecystitis ( n  = 3; 5.9%). Most of the patients had multiple comorbidities; these were diabetes mellitus, hypertension, cardiovascular disease, chronic renal disease, underlying malignancy, and multisystem disease with sepsis. All patients had undergone PC through transhepatic approach under ultrasound guidance in ICU. Technical success rate of the procedure was 100%. Clinical success rate was 92.1% (47/51) and among these 44/51 (86.2%) patients underwent definitive elective cholecystectomy, 3/51 (5.9%) patients had elective tube removal. Three of fifty-one (5.9%) patients did not improve; among these two underwent emergency surgery, while there was 1/51 (1.9%) mortality due to ongoing sepsis and multiorgan dysfunction. There were no procedure-related mortalities or procedure-related major complications. One patient had bile leak due to multiple attempts for cholecystostomy placement. Mean tube indwelling time was 13 days (range: 3–45 days). Conclusion  Ultrasound-guided PC can be safely performed in ICU in critically ill patients unfit for surgery with high technical and clinical success rates. Early laparoscopic cholecystectomy should be preferred after stabilization of clinical condition following cholecystostomy.","PeriodicalId":51597,"journal":{"name":"Indian Journal of Radiology and Imaging","volume":"325 9","pages":""},"PeriodicalIF":0.6,"publicationDate":"2023-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139152442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}