Maxime Pastor , Djamel Dabli , Raphaël Lonjon , Chris Serrand , Fehmi Snene , Fayssal Trad , Fabien de Oliveira , Jean-Paul Beregi , Joël Greffier
{"title":"Comparison between artificial intelligence solution and radiologist for the detection of pelvic, hip and extremity fractures on radiographs in adult using CT as standard of reference","authors":"Maxime Pastor , Djamel Dabli , Raphaël Lonjon , Chris Serrand , Fehmi Snene , Fayssal Trad , Fabien de Oliveira , Jean-Paul Beregi , Joël Greffier","doi":"10.1016/j.diii.2024.09.004","DOIUrl":"10.1016/j.diii.2024.09.004","url":null,"abstract":"<div><h3>Purpose</h3><div>The purpose of this study was to compare the diagnostic performance of an artificial intelligence (AI) solution for the detection of fractures of pelvic, proximal femur or extremity fractures in adults with radiologist interpretation of radiographs, using standard dose CT examination as the standard of reference.</div></div><div><h3>Materials and methods</h3><div>This retrospective study included 94 adult patients with suspected bone fractures who underwent a standard dose CT examination and radiographs of the pelvis and/or hip and extremities at our institution between January 2022 and August 2023. For all patients, an AI solution was used retrospectively on the radiographs to detect and localize bone fractures of the pelvis and/or hip and extremities. Results of the AI solution were compared to the reading of each radiograph by a radiologist using McNemar test. The results of standard dose CT examination as interpreted by a senior radiologist were used as the standard of reference.</div></div><div><h3>Result</h3><div>A total of 94 patients (63 women; mean age, 56.4 ± 22.5 [standard deviation] years) were included. Forty-seven patients had at least one fracture, and a total of 71 fractures were deemed present using the standard of reference (25 hand/wrist, 16 pelvis, 30 foot/ankle). Using the standard of reference, the analysis of radiographs by the AI solution resulted in 58 true positive, 13 false negative, 33 true negative and 15 false positive findings, yielding 82 % sensitivity (58/71; 95 % confidence interval [CI]: 71–89 %), 69 % specificity (33/48; 95 % CI: 55–80 %), and 76 % accuracy (91/119; 95 % CI: 69–84 %). Using the standard of reference, the reading of the radiologist resulted in 65 true positive, 6 false negative, 42 true negative and 6 false positive findings, yielding 92 % sensitivity (65/71; 95 % CI: 82–96 %), 88 % specificity (42/48; 95 % CI: 75–94 %), and 90 % accuracy (107/119; 95 % CI: 85–95 %). The radiologist outperformed the AI solution in terms of sensitivity (<em>P</em> = 0.045), specificity (<em>P</em> = 0.016), and accuracy (<em>P</em> < 0.001).</div></div><div><h3>Conclusion</h3><div>In this study, the radiologist outperformed the AI solution for the diagnosis of pelvic, hip and extremity fractures of the using radiographs. This raises the question of whether a strong standard of reference for evaluating AI solutions should be used in future studies comparing AI and human reading in fracture detection using radiographs.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"106 1","pages":"Pages 22-27"},"PeriodicalIF":4.9,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142298957","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}
{"title":"Artificial intelligence in interventional radiology: Current concepts and future trends","authors":"Armelle Lesaunier , Julien Khlaut , Corentin Dancette , Lambros Tselikas , Baptiste Bonnet , Tom Boeken","doi":"10.1016/j.diii.2024.08.004","DOIUrl":"10.1016/j.diii.2024.08.004","url":null,"abstract":"<div><div>While artificial intelligence (AI) is already well established in diagnostic radiology, it is beginning to make its mark in interventional radiology. AI has the potential to dramatically change the daily practice of interventional radiology at several levels. In the preoperative setting, recent advances in deep learning models, particularly foundation models, enable effective management of multimodality and increased autonomy through their ability to function minimally without supervision. Multimodality is at the heart of patient-tailored management and in interventional radiology, this translates into the development of innovative models for patient selection and outcome prediction. In the perioperative setting, AI is manifesting itself in applications that assist radiologists in image analysis and real-time decision making, thereby improving the efficiency, accuracy, and safety of interventions. In synergy with advances in robotic technologies, AI is laying the groundwork for an increased autonomy. From a research perspective, the development of artificial health data, such as AI-based data augmentation, offers an innovative solution to this central issue and promises to stimulate research in this area. This review aims to provide the medical community with the most important current and future applications of AI in interventional radiology.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"106 1","pages":"Pages 5-10"},"PeriodicalIF":4.9,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142194565","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}
{"title":"Artificial intelligence for bone fracture detection: A promising tool but no substitute for human expertise","authors":"Daphné Guenoun , Mickaël Tordjman","doi":"10.1016/j.diii.2024.10.004","DOIUrl":"10.1016/j.diii.2024.10.004","url":null,"abstract":"","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"106 1","pages":"Pages 3-4"},"PeriodicalIF":4.9,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142510986","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}
{"title":"Artificial intelligence in radiotherapy: Current applications and future trends","authors":"Paul Giraud , Jean-Emmanuel Bibault","doi":"10.1016/j.diii.2024.06.001","DOIUrl":"10.1016/j.diii.2024.06.001","url":null,"abstract":"<div><div>Radiation therapy has dramatically changed with the advent of computed tomography and intensity modulation. This added complexity to the workflow but allowed for more precise and reproducible treatment. As a result, these advances required the accurate delineation of many more volumes, raising questions about how to delineate them, in a uniform manner across centers. Then, as computing power improved, reverse planning became possible and three-dimensional dose distributions could be generated. Artificial intelligence offers the opportunity to make such workflow more efficient while increasing practice homogeneity. Many artificial intelligence-based tools are being implemented in routine practice to increase efficiency, reduce workload and improve homogeneity of treatments. Data retrieved from this workflow could be combined with clinical data and omic data to develop predictive tools to support clinical decision-making process. Such predictive tools are at the stage of proof-of-concept and need to be explainatory, prospectively validated, and based on large and multicenter cohorts. Nevertheless, they could bridge the gap to personalized radiation oncology, by personalizing oncologic strategies, dose prescriptions to tumor volumes and dose constraints to organs at risk.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"105 12","pages":"Pages 475-480"},"PeriodicalIF":4.9,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141451976","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}
{"title":"Diagnostic performance and relationships of structural parameters and strain components for the diagnosis of cardiac amyloidosis with MRI","authors":"Youssef Zaarour , Islem Sifaoui , Haifa Remili , Mounira Kharoubi , Amira Zaroui , Thibaud Damy , Jean-François Deux","doi":"10.1016/j.diii.2024.08.002","DOIUrl":"10.1016/j.diii.2024.08.002","url":null,"abstract":"<div><h3>Purpose</h3><div>The purpose of this study was to evaluate the diagnostic performance and relationships of cardiac MRI structural parameters and strain components in patients with cardiac amyloidosis (CA) and to estimate the capabilities of these variables to discriminate between CA and non-amyloid cardiac hypertrophy (NACH).</div></div><div><h3>Materials and methods</h3><div>Seventy patients with CA (56 men; mean age, 76 ± 10 [standard deviation] years) and 32 patients (19 men; mean age, 63 ± 10 [standard deviation] years) with NACH underwent cardiac MRI. Feature tracking (FT) global longitudinal strain (GLS), radial strain (GRS), circumferential strain (GCS), strain AB ratio (apical strain divided by basal strain), myocardial T1, myocardial T2 and extracellular volume (ECV) were calculated. Comparisons between patients with CA and those with NACH were made using Mann-Whitney rank sum test. The ability of each variable to discriminate between CA and NACH was estimated using area under the receiver operating characteristic curve (AUC).</div></div><div><h3>Results</h3><div>Patients with CA had higher median GLS (-7.0% [Q1, -9.0; Q3, -5.0]), higher median GCS (-12.0% [Q1, -15.0; Q3, -9.0]), and lower median GRS (16.5% [Q1, 13.0; Q3, 23.0]) than those with NACH (-9.0% [Q1, -11.0; Q3, -8.0]; -17.0% [Q1, -20.0; Q3, -14.0]; and 25.5% [Q1, 16.0; Q3, 31.5], respectively) (<em>P</em> < 0.001 for all). Median myocardial T1 and ECV were significantly higher in patients with CA (1112 ms [Q1, 1074; Q3, 1146] and 47% [Q1, 41; Q3, 55], respectively) than in those with NACH (1056 ms [Q1, 1011; Q3, 1071] and 28% [Q1, 26; Q3, 30], respectively) (<em>P</em> < 0.001). Basal ECV showed the best performance for the diagnosis of CA (AUC = 0.975; 95% confidence interval [CI]: 0.947–1). No differences in AUC were found between AB ratio of GRS (0.843; 95% CI: 0.768–0.918) and basal myocardial T1 (0.834; 95% CI: 0.741–0.928) for the diagnosis of CA (<em>P</em> = 0.81). The combination of the AB ratio of FT-GRS and basal myocardial T1 had a diagnostic performance not different from that of basal ECV (<em>P</em> = 0.06).</div></div><div><h3>Conclusion</h3><div>ECV outperforms FT-strain for the diagnosis of CA with cardiac MRI. The AB ratio of FT-GRS associated with myocardial T1 provides diagnostic performance similar to that achieved by ECV.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"105 12","pages":"Pages 489-497"},"PeriodicalIF":4.9,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142134256","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}
Flora Lamant , Gabriel Simon , Andreas Busse-Coté , Youness Hassoun , Bastien Roussel , Pierre Verdot , Alexandre Doussot , Zaher Lakkis , Eric Delabrousse , Paul Calame
{"title":"Assessment of small bowel ischemia in mechanical small bowel obstruction: Diagnostic value of bowel wall iodine concentration using dual-energy CT","authors":"Flora Lamant , Gabriel Simon , Andreas Busse-Coté , Youness Hassoun , Bastien Roussel , Pierre Verdot , Alexandre Doussot , Zaher Lakkis , Eric Delabrousse , Paul Calame","doi":"10.1016/j.diii.2024.10.009","DOIUrl":"10.1016/j.diii.2024.10.009","url":null,"abstract":"<div><h3>Purpose</h3><div>The purpose of this study was to determine whether dual-energy computed tomography (DECT), specifically by measuring bowel wall iodine concentration (BWIC), is superior to monoenergetic reconstructions (MR) for the diagnosis and staging of small bowel ischemia in patients with mechanical small bowel obstruction (SBO).</div></div><div><h3>Materials and methods</h3><div>From November 2021 to December 2023, all patients with mechanical SBO who underwent contrast-enhanced DECT of the abdomen and pelvis were evaluated for inclusion. Demographic, clinical and biochemical data were collected. Two abdominal radiologists, blinded to all patient information, reviewed all DECT examinations. Conventional CT features (including a closed loop mechanism, mesenteric haziness, decreased bowel wall enhancement (DBE), and increased unenhanced attenuation of the bowel wall) were first reviewed on 70-keV-MR and 40-keV-MR, followed by BWIC measurements in five regions of interest in the walls of both normal and abnormal small bowel loops. The diagnostic performance of a simple CT score, which included a closed loop mechanism, mesenteric haziness and DBE, was compared to that of BWIC measurements made on dilated and/or abnormal small bowel segments for the diagnosis of small bowel ischemia. The diagnostic capabilities were compared using areas under the receiver operating characteristic curves (AUCs).</div></div><div><h3>Results</h3><div>A total of 142 patients were included (80 men, 62 women; mean age, 67 ± 17 [standard deviation (SD)] years). Fifty-six patients underwent surgery; 22 of them had confirmed small bowel ischemia, including 12 patients with small bowel necrosis requiring surgical resection. Significant differences in mean BWIC were found between patients without small bowel ischemia (1.73 ± 0.44 [SD] mg/mL), those with small bowel ischemia without necrosis (0.79 ± 0.37 [SD] mg/mL), and those with small bowel ischemia and necrosis (0.48 ± 0.32 [SD] mg/mL) (<em>P</em> < 0.001). The overall AUC of the BWIC measurement for diagnosing small bowel ischemia was 0.98 (95 % confidence interval [CI]: 0.97–1.00), similar to the AUC of the simple CT score (0.97; 95 % CI: 0.92–1.00). However, using a cut off-value of 1.16 mgI/mL, BWIC outperformed subjective assessment of DBE at 70-keV-MR and 40-keV-MR (Youden index, 0.90 vs. 0.54 and vs. 0.71, respectively) (<em>P</em> < 0.001 for both).</div></div><div><h3>Conclusion</h3><div>BWIC measurement outperforms subjective assessment of DBE for the diagnosis of small bowel ischemia in patients with SBO and can allow stratification of ischemia. However, BWIC does not outperfomr a global comprehensive analysis of conventional CT images.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"106 4","pages":"Pages 126-134"},"PeriodicalIF":4.9,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142644096","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}
{"title":"Differentiating neoplastic from bland portal vein thrombus using dual-energy CT","authors":"Bastien Roussel , Gabriel Simon , Paul Calame","doi":"10.1016/j.diii.2024.10.008","DOIUrl":"10.1016/j.diii.2024.10.008","url":null,"abstract":"","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"106 4","pages":"Pages 115-116"},"PeriodicalIF":4.9,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142630673","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}
Cécile Masson-Grehaigne , Mathilde Lafon , Jean Palussière , Laura Leroy , Benjamin Bonhomme , Eva Jambon , Antoine Italiano , Sophie Cousin , Amandine Crombé
{"title":"Single- and multi-site radiomics may improve overall survival prediction for patients with metastatic lung adenocarcinoma","authors":"Cécile Masson-Grehaigne , Mathilde Lafon , Jean Palussière , Laura Leroy , Benjamin Bonhomme , Eva Jambon , Antoine Italiano , Sophie Cousin , Amandine Crombé","doi":"10.1016/j.diii.2024.07.005","DOIUrl":"10.1016/j.diii.2024.07.005","url":null,"abstract":"<div><h3>Purpose</h3><div>The purpose of this study was to assess whether single-site and multi-site radiomics could improve the prediction of overall survival (OS) of patients with metastatic lung adenocarcinoma compared to clinicopathological model.</div></div><div><h3>Materials and methods</h3><div>Adults with metastatic lung adenocarcinoma, pretreatment whole-body contrast-enhanced computed tomography examinations, and performance status (WHO-PS) ≤ 2 were included in this retrospective single-center study, and randomly assigned to training and testing cohorts. Radiomics features (RFs) were extracted from all measurable lesions with volume ≥ 1 cm<sup>3</sup>. Radiomics prognostic scores based on the largest tumor (RPS<sub>largest</sub>) and the average RF values across all tumors per patient (RPS<sub>average</sub>) were developed in the training cohort using 5-fold cross-validated LASSO-penalized Cox regression. Intra-patient inter-tumor heterogeneity (IPITH) metrics were calculated to quantify the radiophenotypic dissimilarities among all tumors within each patient. A clinicopathological model was built in the training cohort using stepwise Cox regression and enriched with combinations of RPS<sub>average</sub>, RPS<sub>largest</sub> and IPITH. Models were compared with the concordance index in the independent testing cohort.</div></div><div><h3>Results</h3><div>A total of 300 patients (median age: 63.7 years; 40.7% women; median OS, 16.3 months) with 1359 lesions were included (200 and 100 patients in the training and testing cohorts, respectively). The clinicopathological model included WHO-PS = 2 (hazard ratio [HR] = 3.26; <em>P</em> < 0.0001), EGFR, ALK, ROS1 or RET mutations (HR = 0.57; <em>P =</em> 0.0347), IVB stage (HR = 1.65; <em>P =</em> 0.0211), and liver metastases (HR = 1.47; <em>P =</em> 0.0670). In the testing cohort, RPS<sub>average</sub>, RPS<sub>largest</sub> and IPITH were associated with OS (HR = 85.50, <em>P =</em> 0.0038; HR = 18.83, <em>P =</em> 0.0082 and HR = 8.00, <em>P =</em> 0.0327, respectively). The highest concordance index was achieved with the combination of clinicopathological variables and RPS<sub>average</sub>, significantly better than that of the clinicopathological model (concordance index = 0.7150 vs. 0.695, respectively; <em>P =</em> 0.0049)</div></div><div><h3>Conclusion</h3><div>Single-site and multi-site radiomics-based scores are associated with OS in patients with metastatic lung adenocarcinoma. RPS<sub>average</sub> improves the clinicopathological model.</div></div>","PeriodicalId":48656,"journal":{"name":"Diagnostic and Interventional Imaging","volume":"105 11","pages":"Pages 439-452"},"PeriodicalIF":4.9,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142082364","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}