Acta radiologica最新文献

{"title":"Deep-learning reconstruction enhances image quality of Adamkiewicz Artery in low-keV dual-energy CT.","authors":"Fuminari Tatsugami, Toru Higaki, Ikuo Kawashita, Chikako Fujioka, Yuko Nakamura, Shinya Takahashi, Kazuo Awai","doi":"10.1177/02841851241288507","DOIUrl":"https://doi.org/10.1177/02841851241288507","url":null,"abstract":"<p><strong>Background: </strong>Low-keV virtual monoenergetic images (VMIs) of dual-energy computed tomography (CT) enhances iodine contrast for detecting small arteries like the Adamkiewicz artery (AKA), but image noise can be problematic. Deep-learning image reconstruction (DLIR) effectively reduces noise without sacrificing image quality.</p><p><strong>Purpose: </strong>To evaluate whether DLIR on low-keV VMIs of dual-energy CT scans improves the visualization of the AKA.</p><p><strong>Material and methods: </strong>We enrolled 29 patients who underwent CT angiography before aortic repair. VMIs obtained at 70 and 40 keV were reconstructed using hybrid iterative reconstruction (HIR), and 40 keV VMIs were reconstructed using DLIR. The image noise of the spinal cord, the maximum CT values of the anterior spinal artery (ASA), and the contrast-to-noise ratio (CNR) of the ASA were compared. The overall image quality and the delineation of the AKA were evaluated on a 4-point score (1 = poor, 4 = excellent).</p><p><strong>Results: </strong>The mean image noise of the spinal cord was significantly lower on 40-keV DLIR than on 40-keV HIR scans; they were significantly higher than on 70-keV HIR images. The CNR of the ASA was highest on the 40-keV DLIR images among the three reconstruction images. The mean image quality scores for 40-keV DLIR and 70-keV HIR scans were comparable, and higher than of 40-keV HIR images. The mean delineation scores for 40-keV HIR and 40-keV DLIR scans were significantly higher than for 70-keV HIR images.</p><p><strong>Conclusion: </strong>Visualization of the AKA was significantly better on low-keV VMIs subjected to DLIR than conventional HIR images.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"2841851241288507"},"PeriodicalIF":1.1,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142455563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The association between ultrasonographic findings and pain, physical function, and symptoms in patients with shoulder pain.","authors":"Onur Engin, Atilla Hikmet Çilengir, Berna Dirim Mete","doi":"10.1177/02841851241287309","DOIUrl":"https://doi.org/10.1177/02841851241287309","url":null,"abstract":"<p><strong>Background: </strong>Shoulder ultrasonography (US) is commonly employed to differentiate the causes of shoulder pain. However, the correlation between ultrasonographic findings and the individual's functional disability and daily activities remains unexplored.</p><p><strong>Purpose: </strong>To investigate the relationship between shoulder US findings and physical function, disability, and pain.</p><p><strong>Material and methods: </strong>The study investigated patients with shoulder pain. Tendons of supraspinatus, infraspinatus, subscapularis, and long head of biceps brachii, acromiohumeral distance, and acromioclavicular joint (ACJ) degeneration were evaluated using US. The QuickDASH questionnaire was employed to evaluate physical function, symptoms, and disability, while the visual analog scale (VAS) was utilized to assess pain levels.</p><p><strong>Results: </strong>The study included 84 patients (69 [82.1%] women, mean age = 55.7 ± 11.9 years). Ultrasonographic examination revealed ACJ degeneration in 61 (21.4%) patients and biceps tendinitis in 21 (25%) patients. In addition, supraspinatus tendinosis was identified in 51 (61%) patients, while 28 (33%) exhibited a supraspinatus tear. The patients demonstrated a mean VAS score of 7.03 ± 1.49 and mean QuickDASH score of 45.6 ± 15.3. Individuals with a supraspinatus tear exhibited higher VAS (<i>P</i> = 0.008) and QuickDASH (<i>P</i> = 0.016) scores. Patients presenting with ACJ degeneration demonstrated increased VAS scores (<i>P</i> = 0.014), whereas those with biceps tendinopathy showed higher QuickDASH scores (<i>P</i> = 0.035).</p><p><strong>Conclusion: </strong>The results obtained from our research demonstrate significant correlations between ultrasonographic data and pain, disability, and physical manifestations. The findings of this study indicate that shoulder US provides valuable information regarding the difficulties individuals face in their daily tasks and the degree of pain they endure.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"2841851241287309"},"PeriodicalIF":1.1,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142455570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fully automated measurement of noise, signal-to-noise ratio, and contrast-to-noise ratio on chest CT images: feasibility and efficiency.","authors":"Bozhe Mei, Zhangman Ma, Wanyun Fu, Linyang He, Zhicheng Ma, Xiangyang Gong","doi":"10.1177/02841851241287315","DOIUrl":"https://doi.org/10.1177/02841851241287315","url":null,"abstract":"<p><strong>Background: </strong>Rapid and accurate measurement of computed tomography (CT) image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) is a clinical challenge.</p><p><strong>Purpose: </strong>To explore the feasibility of intelligent measurement of chest CT image noise, SNR, and CNR.</p><p><strong>Material and methods: </strong>A total of 300 chest CT scans were included in the study, which was divided into research dataset, internal test dataset, and external test dataset. Based on the research dataset, automatically segment and measure the average CT values and standard deviation (SD) of CT values for background air and lung field under different thresholds to obtain noise, SNR, and CNR results. Using the results of manual measurements as the reference standard, we determine the optimal threshold with the highest consistency. Using internal and external test datasets, validate the consistency of automated measurements of noise, SNR, and CNR at the optimal CT threshold with reference standards.</p><p><strong>Results: </strong>With background air set at -900 HU and lung field at -800 HU as thresholds, the automated measurements of noise, SNR, and CNR demonstrate the highest consistency with the reference standards. At the optimal threshold, the noise, SNR, and CNR measured automatically on both the internal (intraclass correlation coefficient [ICC] = 0.85-0.96) and external (ICC = 0.75-0.85) test datasets exhibit high consistency with their respective reference standards.</p><p><strong>Conclusion: </strong>The method we explored can intelligently measure the noise, SNR, and CNR of chest CT images, exhibits high consistency with radiologists, and offers a novel tool for image quality evaluation and analysis.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"2841851241287315"},"PeriodicalIF":1.1,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142455567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of pediatric breast ultrasound less is more: a practical imaging approach.","authors":"Elisabetta Giannotti, Rachel Sun, Nuala Healy, Fleur Kilburn-Toppin, Carmelo Sofia, Andrew Hs Lee, Maria Adele Marino","doi":"10.1177/02841851241287924","DOIUrl":"https://doi.org/10.1177/02841851241287924","url":null,"abstract":"<p><strong>Background: </strong>Breast cancer in pediatric patients is rare, but ultrasound (US) is widely utilized for symptomatic cases.</p><p><strong>Purpose: </strong>To determine biopsy and cancer detection rates of pediatric patients and to assess if breast US can be omitted.</p><p><strong>Material and methods: </strong>A retrospective review of a 5-year period was conducted of single-center breast US performed in patients aged <19 years. Data regarding presentation, clinical opinion (P1-5 score), and US (U1-5 score) were collected. If biopsy or surgery was performed, pathology was reviewed (B1-5 score).</p><p><strong>Results: </strong>In total, 579 patients were included (19 boys, 560 girls; mean age=16.2±1.9 years; age range=0-18 years). Clinical examination was normal or benign (P1/P2) in all boys (100%) and 557/560 (99.5%) girls, and P3 in 3 (0.5%) girls. Of US, 52% demonstrated normal findings (U1) for both sexes (300/579); in the remaining cases, the most frequent findings were gynecomastia in 12/19 boys and well-defined breast masses in 208/560 girls. Of the 560 girls, 6 (1%) underwent US-guided biopsy, with final histology of fibroadenoma (B2) in all cases, while 27 (5%) had a surgical excision, with final histology of fibroadenoma (22/27, 81.5%), hamartoma (2/27, 7.4%), benign phyllodes tumor (2/27, 7.4%), and angiomyxoma skin lesion (1/27, 3.7%). No malignant lesions were diagnosed at the time of clinical referral or during the 18-month follow-up in patients with a well-defined mass on US.</p><p><strong>Conclusion: </strong>Breast malignancy is extremely rare in pediatric population. US can be safely omitted if clinical examination is normal; this approach would have avoided breast US in 52% of patients in this study.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"2841851241287924"},"PeriodicalIF":1.1,"publicationDate":"2024-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142455561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A new era of high-resolution CT diagnostics of the lung: improved image quality, detailed morphology, and reduced radiation dose with high-resolution photon-counting CT of the lungs compared to high-resolution energy-integrated CT.","authors":"Marie-Louise Aurumskjöld, Lotta Sjunnesson, Adrian Pistea, Gylfi Ásbjörnsson, Fredrik Wellman, Gracijela Bozovic","doi":"10.1177/02841851241269918","DOIUrl":"10.1177/02841851241269918","url":null,"abstract":"<p><strong>Background: </strong>High-resolution computed tomography (HRCT) is dependent on detailed morphology in diagnostic assessment of interstitial lung diseases. Photon-counting CT (PCCT) enables improved resolution while reducing radiation.</p><p><strong>Purpose: </strong>To compare if the image quality, detailed morphology, and radiation dose in HRCT of the lung improves with PCCT compared to energy-integrated CT (EICT).</p><p><strong>Material and methods: </strong>HRCT with PCCT in patients with body mass index (BMI) from normal to obese, previously examined with different EICT were included. They were evaluated in a five-step scale for image quality according to Quality Criteria for CT (Diagnostic Requirement of the ImPACT group-European standardization). In addition, ground-glass opacities, bronchiectasis, emphysema, nodules, and subpleural detailed morphology (≤1 cm from the pleural border) were evaluated by three independent thoracic and/or pediatric radiologists. Visual grading characteristics (VGC) were used for comparison of image quality and detailed morphology and Fleiss kappa for intra-observer variability. Dose-length product (DLP) and CT dose index-volume (CTDI_vol) were collected to calculate effective radiation dose.</p><p><strong>Results: </strong>HRCT with PCCT in 52 women and 48 men (mean age=67.2 ± 13.6 years; age range=27-87 years; BMI=26.9 kg/m²; range=18.6-45 kg/m²) previously examined with EICT (mean age=65.3 ± 13.6 years; age range=27-85 years; BMI=27 kg/m²; range=18.9-45 kg/m²) were included. There were significant differences in image quality for all entities in favor of PCCT. The radiation dose was reduced with PCCT by 47% in all, particularly pronounced in obese with 48.5%.</p><p><strong>Conclusion: </strong>Image quality, detailed morphology, and radiation dose, particularly in obese patients, were significantly improved in HRCT with PCCT compared to conventional EICT. The new technique enables visualization of subpleural structures.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"1211-1221"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142003371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting axillary lymph node metastasis in breast cancer based on ultrasound radiofrequency time-series analysis.","authors":"Pengfei Sun, Ruifang Guo, Xiangdong Hu, Andre Dekker, Alberto Traverso, Linxue Qian, Zhixiang Wang","doi":"10.1177/02841851241268463","DOIUrl":"10.1177/02841851241268463","url":null,"abstract":"<p><strong>Background: </strong>The status of axillary lymph nodes (ALN) plays a critical role in the management of patients with breast cancer. It is an urgent demand to develop highly accurate, non-invasive methods for predicting ALN status.</p><p><strong>Purpose: </strong>To evaluate the efficacy of ultrasound radiofrequency (URF) time-series parameters, in combination with clinical data, in predicting ALN metastasis in patients with breast cancer.</p><p><strong>Material and methods: </strong>We prospectively gathered clinicopathologic and ultrasonic data from patients diagnosed with breast cancer. Various machine-learning (ML) models were developed using all available features to determine the most efficient diagnostic model. Subsequently, distinct prediction models were created using the optimal ML model, and their diagnostic performances were evaluated and compared.</p><p><strong>Results: </strong>The study encompassed 240 patients, of whom 88 had lymph node metastases. A leave-one-out cross-validation (LOOCV) method was used to split the entire dataset into training and testing subsets. The random forest ML model outperformed the other algorithms, with an area under the curve (AUC) of 0.92. Prediction models based on clinical, ultrasonic, URF parameters, clinical + ultrasonic, clinical + URF, and ultrasonic + URF parameters had AUCs of 0.56, 0.79, 0.78, 0.90, 0.80, and 0.84, respectively, in the testing set. The comprehensive diagnostic model (clinical + ultrasonic + URF parameters) demonstrated strong diagnostic capability, with an AUC of 0.94 in the testing set, exceeding any single prediction model.</p><p><strong>Conclusion: </strong>The combined model (clinical + ultrasonic + URF parameters) could be used preoperatively to predict lymph node status, offering valuable input for the design of individualized surgical approaches.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"1178-1185"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142103323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Increased extracellular free water is related to white matter hyperintensity burden.","authors":"Xinjun Lei, Weiwen Qiu, Zhihua Xu, Jie Yu, Haiyuan Lan","doi":"10.1177/02841851241282085","DOIUrl":"10.1177/02841851241282085","url":null,"abstract":"<p><strong>Background: </strong>Extracellular free water (FW) has important roles in the occurrence and development of white matter hyperintensity (WMH).</p><p><strong>Purpose: </strong>To explore the correlations between FW and WMH burden.</p><p><strong>Material and methods: </strong>A prospective analysis was conducted using magnetic resonance imaging (MRI) data from 126 individuals. WMH burden was determined based on WMH volumes and Fazekas scores from deep and periventricular white matter hyperintensity (DWMH and PWMH, respectively) in fluid-attenuated inversion recovery (FLAIR) images. FW values were taken from diffusion tensor imaging (DTI).</p><p><strong>Results: </strong>Univariate analysis showed that FW values were correlated with WMH burden, including WMH volumes and DWMH and PWMH Fazekas scores (<i>P </i>< 0.05). After multivariate analysis, FW values were correlated with WMH volumes and DWMH and PWMH Fazekas scores when adjusted for age and hypertension (<i>P </i>< 0.05).</p><p><strong>Conclusion: </strong>Using MRI, increasing extracellular FW was related to WMH burden.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"1265-1271"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142278723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Peak enhancement in contrast-enhanced ultrasound for assessing endoscopic disease activity in patients with Crohn's disease: a meta-analysis.","authors":"Lei Yuan, Ruifang Hao, Leilei Zou, Deyin Zhai, Junlian Wang, Ruining Jiao","doi":"10.1177/02841851241275982","DOIUrl":"10.1177/02841851241275982","url":null,"abstract":"<p><strong>Background: </strong>Contrast-enhanced ultrasound (CEUS) has been proposed as a valuable tool for detecting disease activity in patients with Crohn's disease (CD). However, previous studies have utilized different parameters, leading to variation in clinical assessment of this technique.</p><p><strong>Purpose: </strong>To assess the effectiveness of peak enhancement (PE) in CEUS for evaluating endoscopic disease activity in patients with CD.</p><p><strong>Material and methods: </strong>Articles were obtained by searching PubMed, Embase, Web of Science, Wanfang, and CNKI databases. Only studies that investigated the effectiveness of PE in CEUS to discriminate endoscopic disease activity in patients with CD were considered. Pooled sensitivity and specificity were then calculated using a random effects model.</p><p><strong>Results: </strong>Overall, seven studies were included. The endoscopic disease activity of CD was determined based on the simple endoscopic score for Crohn's disease and Crohn's Disease Endoscopic Index of Severity scores. Pooled results showed that a high PE was associated with increased detection efficacy for endoscopic disease activity in CD. Pooled sensitivity, specificity, and positive and negative likelihood ratios were 0.88 (95% confidence interval [CI] = 0.71-0.96), 0.88 (95% CI = 0.81-0.93), 7.60 (95% CI = 4.61-12.53), and 0.14 (95% CI = 0.05-0.35), respectively. The pooled receiver operating characteristic was 0.90 (95% CI = 0.87-0.92), suggesting a good discriminating efficacy of PE in CEUS for endoscopic disease activity of patients with CD.</p><p><strong>Conclusions: </strong>A high PE in CEUS displayed substantial distinguishing accuracy for assessing endoscopic disease activity of patients with CD.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"1170-1177"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142278725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiographic assessment of incidental bone lesions of the proximal humerus: a prevalence study.","authors":"Mustafa Kara, Serkan Bayram, Emre Kocazeybek, Gökhan Demirtaş","doi":"10.1177/02841851241275058","DOIUrl":"10.1177/02841851241275058","url":null,"abstract":"<p><strong>Background: </strong>An increased prevalence of benign lesions has been associated with the increased use of radiological tools in orthopedic practice.</p><p><strong>Purpose: </strong>To investigate the inter-observer reliability of the radiographic detection of lesions on the proximal humerus between different observers.</p><p><strong>Material and methods: </strong>X-ray radiological examinations of 17,243 patients were performed by an orthopedic surgeon to assess the presence of proximal humeral bone lesions. After this initial screening, 201 patients with lesions were identified and these 201 scans were assessed by three different physicians: an orthopedic oncologist, an orthopedic surgeon, and a musculoskeletal radiologist. Of the 201 patients with lesions, 80 had magnetic resonance imaging in addition to X-rays and were assessed. The diagnosis of the lesions was determined by each participant according to their radiological appearance to assess inter-observer reliability.</p><p><strong>Results: </strong>The prevalence of proximal humeral lesions was determined by the orthopedic surgeon to be 1.165% (201 of 17,243 patients). In 201 patients with a lesion, the mean age was 48 ± 17 years (range = 18-91 years). According to the assessments of the orthopedic oncologist, orthopedic surgeon, and musculoskeletal radiologist, the most common bone lesion identified was enostosis, with prevalence percentages of 53.7%, 59.7%, and 64.7%, respectively. A noteworthy consensus exists in the collective diagnosis of bone lesions among the three investigators, demonstrated by a substantial agreement with Fleiss' κ at 0.74 (<i>P</i> < 0.001).</p><p><strong>Conclusion: </strong>There was remarkable agreement between the three investigators on the collective diagnosis of bone lesions. Enostosis was the most common bone lesion at the proximal humerus in our series.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"1246-1251"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142103324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning and radiomics for ventricular tachyarrhythmia prediction in hypertrophic cardiomyopathy: insights from an MRI-based analysis.","authors":"Emine Sebnem Durmaz, Mert Karabacak, Burak Berksu Ozkara, Osman Aykan Kargın, Bilal Demir, Damla Raimoglou, Ahmet Atil Aygun, Ibrahim Adaletli, Ahmet Bas, Eser Durmaz","doi":"10.1177/02841851241283041","DOIUrl":"https://doi.org/10.1177/02841851241283041","url":null,"abstract":"<p><strong>Background: </strong>Myocardial fibrosis is often detected in patients with hypertrophic cardiomyopathy (HCM), which causes left ventricular (LV) dysfunction and tachyarrhythmias.</p><p><strong>Purpose: </strong>To evaluate the potential value of a machine learning (ML) approach that uses radiomic features from late gadolinium enhancement (LGE) and cine images for the prediction of ventricular tachyarrhythmia (VT) in patients with HCM.</p><p><strong>Material and methods: </strong>Hyperenhancing areas of LV myocardium on LGE images were manually segmented, and the segmentation was propagated to corresponding areas on cine images. Radiomic features were extracted using the PyRadiomics library. The least absolute shrinkage and selection operator (LASSO) method was employed for radiomic feature selection. Our model development employed the TabPFN algorithm, an adapted Prior-Data Fitted Network design. Model performance was evaluated graphically and numerically over five-repeat fivefold cross-validation. SHapley Additive exPlanations (SHAP) were employed to determine the relative importance of selected radiomic features.</p><p><strong>Results: </strong>Our cohort consisted of 60 patients with HCM (73.3% male; median age = 51.5 years), among whom 17 had documented VT during the follow-up. A total of 1612 radiomic features were extracted for each patient. The LASSO algorithm led to a final selection of 18 radiomic features. The model achieved a mean area under the receiver operating characteristic curve of 0.877, demonstrating good discrimination, and a mean Brier score of 0.119, demonstrating good calibration.</p><p><strong>Conclusion: </strong>Radiomics-based ML models are promising for predicting VT in patients with HCM during the follow-up period. Developing predictive models as clinically useful decision-making tools may significantly improve risk assessment and prognosis.</p>","PeriodicalId":7143,"journal":{"name":"Acta radiologica","volume":" ","pages":"2841851241283041"},"PeriodicalIF":1.1,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142339126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}