Research in diagnostic and interventional imaging最新文献

{"title":"Elbow trauma in children: development and evaluation of radiological artificial intelligence models","authors":"Clémence ROZWAG ,&nbsp;Franck VALENTINI ,&nbsp;Anne COTTEN ,&nbsp;Xavier DEMONDION ,&nbsp;Philippe PREUX ,&nbsp;Thibaut JACQUES","doi":"10.1016/j.redii.2023.100029","DOIUrl":"10.1016/j.redii.2023.100029","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><p>To develop a model using artificial intelligence (A.I.) able to detect post-traumatic injuries on pediatric elbow X-rays then to evaluate its performances in silico and its impact on radiologists’ interpretation in clinical practice.</p></div><div><h3>Material and Methods</h3><p>A total of 1956 pediatric elbow radiographs performed following a trauma were retrospectively collected from 935 patients aged between 0 and 18 years. Deep convolutional neural networks were trained on these X-rays. The two best models were selected then evaluated on an external test set involving 120 patients, whose X-rays were performed on a different radiological equipment in another time period. Eight radiologists interpreted this external test set without then with the help of the A.I. models .</p></div><div><h3>Results</h3><p>Two models stood out: model 1 had an accuracy of 95.8% and an AUROC of 0.983 and model 2 had an accuracy of 90.5% and an AUROC of 0.975. On the external test set, model 1 kept a good accuracy of 82.5% and AUROC of 0.916 while model 2 had a loss of accuracy down to 69.2% and of AUROC to 0.793. Model 1 significantly improved radiologist's sensitivity (0.82 to 0.88, <em>P</em> = 0.016) and accuracy (0.86 to 0.88, <em>P</em> = 0,047) while model 2 significantly decreased specificity of readers (0.86 to 0.83, <em>P</em> = 0.031).</p></div><div><h3>Conclusion</h3><p>End-to-end development of a deep learning model to assess post-traumatic injuries on elbow X-ray in children was feasible and showed that models with close metrics in silico can unpredictably lead radiologists to either improve or lower their performances in clinical settings.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"6 ","pages":"Article 100029"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49070855","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":"BERT-based natural language processing analysis of French CT reports: Application to the measurement of the positivity rate for pulmonary embolism","authors":"Émilien Jupin-Delevaux ,&nbsp;Aissam Djahnine ,&nbsp;François Talbot ,&nbsp;Antoine Richard ,&nbsp;Sylvain Gouttard ,&nbsp;Adeline Mansuy ,&nbsp;Philippe Douek ,&nbsp;Salim Si-Mohamed ,&nbsp;Loïc Boussel","doi":"10.1016/j.redii.2023.100027","DOIUrl":"10.1016/j.redii.2023.100027","url":null,"abstract":"<div><h3>Rationale and objectives</h3><p>To develop a Natural Language Processing (NLP) method based on Bidirectional Encoder Representations from Transformers (BERT) adapted to French CT reports and to evaluate its performance to calculate the diagnostic yield of CT in patients with clinical suspicion of pulmonary embolism (PE).</p></div><div><h3>Materials and methods</h3><p>All the CT reports performed in our institution in 2019 (99,510 reports, training and validation dataset) and 2018 (94,559 reports, testing dataset) were included after anonymization. Two BERT-based NLP sentence classifiers were trained on 27.700, manually labeled, sentences from the training dataset. The first one aimed to classify the reports’ sentences into three classes (“Non chest”, “Healthy chest”, and \"Pathological chest\" related sentences), the second one to classify the last class into eleven sub classes pathologies including \"pulmonary embolism\". F1-score was reported on the validation dataset. These NLP classifiers were then applied to requested CT reports for pulmonary embolism from the testing dataset. Sensitivity, specificity, and accuracy for detection of the presence of a pulmonary embolism were reported in comparison to human analysis of the reports.</p></div><div><h3>Results</h3><p>The F1-score for the 3-Classes and 11-SubClasses classifiers was 0.984 and 0.985, respectively. 4,042 examinations from the testing dataset were requested for pulmonary embolism of which 641 (15.8%) were positively evaluated by radiologists. The sensitivity, specificity, and accuracy of the NLP network for identifying pulmonary embolism in these reports were 98.2%, 99.3% and 99.1%, respectively.</p></div><div><h3>Conclusion</h3><p>BERT-based NLP sentences classifier enables the analysis of large databases of radiological reports to accurately determine the diagnostic yield of CT screening.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"6 ","pages":"Article 100027"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44997288","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":"Radiation dose in CT-based active surveillance of small renal masses may be reduced by 75%: A retrospective exploratory multiobserver study","authors":"Jens Borgbjerg ,&nbsp;Nis Elbrønd Larsen ,&nbsp;Ivar Mjåland Salte ,&nbsp;Niklas Revold Grønli ,&nbsp;Elise Klæstrup ,&nbsp;Anne Negård","doi":"10.1016/j.redii.2022.100019","DOIUrl":"10.1016/j.redii.2022.100019","url":null,"abstract":"","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100019"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49336767","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":"(18F)-PSMA-1007PET/CT in patients with biochemical recurrence after radical prostatectomy: Diagnostic performance and impact on treatment management","authors":"Jia Jiang ,&nbsp;Lei Chen ,&nbsp;Xiaowei Ji ,&nbsp;Xuan Zheng ,&nbsp;Junjie Hong ,&nbsp;Kun Tang ,&nbsp;Xiangwu Zheng","doi":"10.1016/j.redii.2022.100021","DOIUrl":"10.1016/j.redii.2022.100021","url":null,"abstract":"<div><h3>Objective</h3><p>To evaluate the diagnostic performance of (¹⁸F)-PSMA-1007 PET/CT in prostate cancer patients with biochemical recurrence (BCR) after radical prostatectomy and the effect of (¹⁸F)-PSMA-1007 PET/CT on treatment strategy.</p></div><div><h3>Methods</h3><p>A total of 114 patients with BCR after radical prostatectomy who performed (¹⁸F)-PSMA-1007 PET/CT were retrospectively analyzed. The Gleason scores (GS), maximum standardized uptake values (SUV_max) and the diagnostic performance were compared according to different prostate-specific antigen (PSA) groups. To evaluate the impact of (¹⁸F)-PSMA-1007 PET/CT on treatment management, we also collected subjects’ therapy before and after PET/CT. The PSA value was monitored to evaluate the biochemical response.</p></div><div><h3>Results</h3><p>(¹⁸F)-PSMA-1007PET/CT was positive in 92/114 patients (80.7%). The detection rates were 20/34 (58.8%), 13/17 (76.5%), 15/17 (88.2%) and 44/46 (95.7%) for PSA levels of 0.2-&lt;0.5, 0.5-&lt;1, 1-&lt;2, ≥2 ng/ml. The positive lesions on PET/CT revealed local recurrence in 24/114 (21.1%) patients, lymph nodes metastases in 54/114 (47.4%) and metastatic sites in bone, lung, and others in 75/114 (65.8%). A significant positive correlation was observed between the GS/ SUV_max and PSA level (r₁ = 0.375, r₂ = 0.336, <em>P</em>&lt;0.001). As a result of the (¹⁸F)-PSMA-1007 PET/CT, therapeutic decision-making changed in 60/114 (52.6%) patients. With a follow-up of 11.0 ± 6.4 months, 81/114 PSA were collected after treatment guided by (¹⁸F)-PSMA-1007 PET/CT, and in 42/81 (51.9%) of patients, serum PSA levels decreased of more than 60%.</p></div><div><h3>Conclusion</h3><p>(¹⁸F)-PSMA-1007 PET/CT has a high lesion detection rate for recurrent prostate cancer (PCa) and could have significant implications in decision-making treatment plan for the majority of PCa patients.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100021"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41492858","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":"Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom","authors":"Tormund Njølstad MD ,&nbsp;Anselm Schulz MD PhD ,&nbsp;Kristin Jensen PhD ,&nbsp;Hilde K. Andersen MSc ,&nbsp;Anne Catrine T. Martinsen PhD","doi":"10.1016/j.redii.2023.100022","DOIUrl":"10.1016/j.redii.2023.100022","url":null,"abstract":"<div><h3>Purpose</h3><p>To assess image quality of a deep learning reconstruction (DLR) algorithm across dose levels using a semi-anthropomorphic upper-abdominal phantom, and compare with filtered back projection (FBP) and hybrid iterative reconstruction (IR).</p></div><div><h3>Material and methods</h3><p>CT scans obtained at five dose levels (CTDI_vol 5, 10, 15, 20 and 25 mGy) were reconstructed with FBP, hybrid IR (IR50, IR70 and IR90) and DLR of low (DLL), medium (DLM) and high strength (DLH) in 0.625 mm and 2.5 mm slices. CT number, homogeneity, noise, contrast, contrast-to-noise ratio (CNR), noise texture deviation (NTD; a measure of IR-specific artifacts), noise power spectrum (NPS) and task-based transfer function (TTF) were compared between reconstruction algorithms.</p></div><div><h3>Results</h3><p>CT numbers were highly consistent across reconstruction algorithms. Image noise was significantly reduced with higher levels of DLR. Noise texture (NPS and NTD) was with DLR maintained at comparable levels to FBP, contrary to increasing levels of hybrid IR. Images reconstructed with DLR of low and high strength in 0.625 mm slices showed similar noise characteristics to 2.5 mm slice FBP and IR50, respectively. Dose-reduction potential based on image noise with IR50 as reference was estimated to 35% for DLM and 74% for DLH.</p></div><div><h3>Conclusions</h3><p>The novel DLR algorithm demonstrates robust noise reduction with maintained noise texture characteristics despite higher algorithm strength, and may have overcome important limitations of IR. There may be potential for dose reduction and additional benefit from thin-slice reconstruction.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100022"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48644660","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":"Dose variations for biopsy, puncture and drainage under CT guidance: A national survey in 1709 patients","authors":"Laure Berny ,&nbsp;Joël Greffier ,&nbsp;Chris Serrand ,&nbsp;Djamel Dabli ,&nbsp;Fabien De Oliveira ,&nbsp;Hélène de Forges ,&nbsp;Jean-Paul Beregi ,&nbsp;Julien Frandon","doi":"10.1016/j.redii.2023.100025","DOIUrl":"10.1016/j.redii.2023.100025","url":null,"abstract":"<div><h3>Background</h3><p>A nation-wide study recently published the dose reference levels for the main CT-guided interventional procedures in 5001 patients. We assessed the impact of patient's age, sex and targeted organ on the patient dose during thoracic and abdominopelvic biopsies and punctures/drainages.</p></div><div><h3>Patients and methods</h3><p>Data were extracted from the previous nationwide study. All biopsies, punctures and drainages for thoracic or abdominopelvic locations performed between January 2017 and June 2019 in all participating centers were included in the study. Multivariable analyses were carried out using a linear regression of the dose-length product (DLP) log, adjusted to age, sex, anatomical location, number of helical acquisitions and inclusion center.</p></div><div><h3>Results</h3><p>Of the 5001 patients of the initial study, 2383 benefited from thoracic or abdominopelvic procedures, including 674 percutaneous destructions excluded. 1709 patients (44 centers), 1045 men, 664 women, median age 64.4 ± 14.0 years were included. The mean DLP was 751.2 ± 642.7mGy.cm. It was significantly higher in men than women (<em>p</em> = 0.0005) and higher for abdominopelvic procedures than for thoracic locations (<em>p</em>&lt;0.0001).</p></div><div><h3>Conclusion</h3><p>Doses delivered to patients for abdominal and thoracic biopsies and punctures/drainages performed under CT guidance depend on gender and location. Furthers studies taking into account the patient's morphology and anatomical location of the procedure would allow proposing finer dose reference levels.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100025"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45522997","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":"Imaging patterns of the arterial supply of the prostate gland in adult Ghanaian men","authors":"Bashiru Babatunde Jimah ,&nbsp;Benjamin Dabo Sarkodie ,&nbsp;Dorothea Anim ,&nbsp;Edmund Brakohiapa ,&nbsp;Asare Kweku Offei ,&nbsp;Ewurama Andam Idun ,&nbsp;Benard Botwe ,&nbsp;Klenam Dzefi-Tettey ,&nbsp;Kofi Amedi","doi":"10.1016/j.redii.2022.100020","DOIUrl":"10.1016/j.redii.2022.100020","url":null,"abstract":"<div><h3>Background</h3><p>Prostatic arterial embolization (PAE) is a novel procedure in West Africa and Ghana. A thorough understanding of the prostate artery's (PA) anatomy and pattern is required for successful prostatic arterial embolization and to guarantee targeted intervention. This study focuses on prostate arterial supply in adult males, including prevalence, variability, and imaging pattern.</p></div><div><h3>Methodology</h3><p>A prospective cross-sectional study was conducted, at Euracare Advanced Diagnostics and Heart Centre. Patients who presented for Computed Tomography Angiography of the pelvis were included in the study. A total of 52 males were included and 104 pelvic CT angiography (one for each side) were analyzed, including: prostatic artery diameter, prostatic gland volume and prostate artery branching pattern. The PA branching pattern was classified using de Assis et al. classification.</p></div><div><h3>Result</h3><p>Thirty-seven (71.15%) men had enlarged prostate volume (&gt;30ml). On each side there was only one prostatic artery and no accessory one was found. Only three types of arterial branching were identified: type I, II,III. The type I artery was the most common origin 58.7% (61/104). PA originating from the anterior division of the internal iliac artery (type II) and the type III is from the internal pudendal artery, accounted for 16.3% (17/104) and 25% (26/104) respectively.</p></div><div><h3>Conclusion</h3><p>The most frequent type of PA origin was type I followed by type III then II. Knowing the different and most frequent types of anatomy of PA may help standardization and effectiveness of the PAE in developing countries.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100020"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45169699","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":"Analysis on diagnostic failure of US-guided core needle biopsy for soft tissue tumors","authors":"Ying-Lun Zhang MS ,&nbsp;Qian Ma MS ,&nbsp;Yu Hu MD ,&nbsp;Meng-Jie Wu MS ,&nbsp;Zong-Kai Wei MS ,&nbsp;Qi-Yu Yao MS ,&nbsp;Ju-Ming Li MD ,&nbsp;Ao Li MD, PhD","doi":"10.1016/j.redii.2023.100023","DOIUrl":"10.1016/j.redii.2023.100023","url":null,"abstract":"<div><h3>Purpose</h3><p>To evaluate the diagnostic yield of ultrasonography (US)-guided core needle biopsy (CNB) in the diagnosis of soft tissue tumors (STTs) and to analyze the failure factors.</p></div><div><h3>Methods</h3><p>139 patients with STTs that underwent both US-guided CNB and surgical resection were collected retrospectively. Compared with the histopathological results of surgical resection, the biopsy failure was defined as the following conditions: indefinitive diagnosis, including insufficient samples and unknown subtypes with correct biological potential classification; wrong diagnosis, including wrong biological potential classification and wrong subtypes with correct biological potential classification. Univariate and multivariate analyses from the perspectives of histopathological, demographic and US features together with biopsy procedures were performed to determine risk factors for diagnostic failure.</p></div><div><h3>Results</h3><p>The diagnostic yield of US-guided CNB for STTs in our study was 78.4%, but when only considering the correct biological potential classification of STTs, the diagnostic yield was 80.6%. The multivariate analysis showed that adipocytic tumors (odds ratio (OR) = 10.195, 95% confidence interval (CI): 1.062 - 97.861, <em>p</em> = 0.044), vascular tumors (OR = 41.710, 95% CI: 3.126 - 556.581, <em>p</em> = 0.005) and indeterminate US diagnosis (OR = 8.641, 95% CI: 1.852 - 40.303, <em>p</em> = 0.006) were correlated with the diagnostic failure. The grade III vascular density (OR = 0.019, 95% CI: 0.001 - 0.273, <em>p</em> = 0.007) enabled a higher diagnostic accuracy.</p></div><div><h3>Conclusion</h3><p>US-guided CNB can be an effective modality for the diagnosis of STTs. The diagnostic yield can be increased when the tumor vascular density was grade III in Color Doppler US, but can be decreased in adipocytic tumors, vascular tumors and masses with indeterminate US diagnosis.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100023"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48004210","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":"Measurement of renal cortical fibrosis by CT scan","authors":"John D Olson ,&nbsp;Janet A Tooze ,&nbsp;Daniel J Bourland ,&nbsp;J Mark Cline ,&nbsp;Eduardo B Faria ,&nbsp;Eric P Cohen","doi":"10.1016/j.redii.2023.100024","DOIUrl":"10.1016/j.redii.2023.100024","url":null,"abstract":"<div><h3>Rationale and objectives</h3><p>The accurate, non-invasive, and rapid measurement of renal cortical fibrosis is needed for well-defined benchmarks of permanent injury and for use of anti-fibrotic agents. It is also needed for non-invasive and rapid assessment of the chronicity of human renal diseases.</p></div><div><h3>Materials and methods</h3><p>We have used a non-human primate model of radiation nephropathy to develop a novel method of size-corrected CT imaging to quantify renal cortical fibrosis.</p></div><div><h3>Results</h3><p>Our method has an area under the receiver operating curve of 0.96, which is superior to any other non-invasive method of measuring renal fibrosis.</p></div><div><h3>Conclusion</h3><p>Our method is suitable for immediate translation to human clinical renal diseases.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"5 ","pages":"Article 100024"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10124964/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9438374","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":"Prediction model for patient prognosis in idiopathic pulmonary fibrosis using hybrid radiomics analysis","authors":"Daisuke Kawahara ,&nbsp;Takeshi Masuda ,&nbsp;Riku Nishioka ,&nbsp;Masashi Namba ,&nbsp;Nobuki Imano ,&nbsp;Kakuhiro Yamaguchi ,&nbsp;Shinjiro Sakamoto ,&nbsp;Yasushi Horimasu ,&nbsp;Shintaro Miyamoto ,&nbsp;Taku Nakashima ,&nbsp;Hiroshi Iwamoto ,&nbsp;Shinichiro Ohshimo ,&nbsp;Kazunori Fujitaka ,&nbsp;Hironobu Hamada ,&nbsp;Noboru Hattori ,&nbsp;Yasushi Nagata","doi":"10.1016/j.redii.2022.100017","DOIUrl":"10.1016/j.redii.2022.100017","url":null,"abstract":"<div><h3>Objectives</h3><p>To develop an imaging prognostic model for idiopathic pulmonary fibrosis (IPF) patients using hybrid auto-segmentation radiomics analysis, and compare the predictive ability between the radiomics analysis and conventional visual score methods.</p></div><div><h3>Methods</h3><p>Data from 72 IPF patients who had undergone CT were analyzed. In the radiomics analysis, quantitative CT analysis was performed using the semi-auto-segmentation method. In the visual method, the extent of radiologic abnormalities was evaluated and the overall percentage of lung involvement was calculated by averaging values for six lung zones. Using a training cohort of 50 cases, we generated a radiomics model and a visual score model. Subsequently, we investigated the predictive ability of these models in a testing cohort of 22 cases.</p></div><div><h3>Results</h3><p>Three significant prognostic factors such as contrast, Idn, and cluster shade were selected by LASSO Cox regression analysis. In the visual method, multivariate Cox regression analysis revealed that honeycombing and reticulation were significant prognostic factors. Subsequently, a predictive nomogram for prognosis in IPF patients was established using these factors. In the testing cohort, the c-index of the visual and radiomics nomograms were 0.68 and 0.74, respectively. When dividing the cohort into high-risk and low-risk groups using the median nomogram score, significant differences in overall survival (OS) in the visual and radiomics models were observed (P=0.000 and P=0.0003, respectively).</p></div><div><h3>Conclusions</h3><p>The prediction model with hybrid radiomics analysis had a better ability to predict OS in IPF patients than that of the visual method.</p></div>","PeriodicalId":74676,"journal":{"name":"Research in diagnostic and interventional imaging","volume":"4 ","pages":"Article 100017"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2772652522000175/pdfft?md5=d46aea44f6d5f4c39c56f34c5d71c7eb&pid=1-s2.0-S2772652522000175-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47231367","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}