Characterization of renal masses with MRI-based radiomics: assessment of inter-package and inter-observer reproducibility in a prospective pilot study.

OBJECTIVES: To evaluate radiomics features' reproducibility using inter-package/inter-observer measurement analysis in renal masses (RMs) based on MRI and to employ machine learning (ML) models for RM characterization. METHODS: 32 Patients (23M/9F; age 61.8 ± 10.6 years) with RMs (25 renal cell carcinomas (RCC)/7 benign masses; mean size, 3.43 ± 1.73 cm) undergoing resection were prospectively recruited. All patients underwent 1.5 T MRI with T2-weighted (T2-WI), diffusion-weighted (DWI)/apparent diffusion coefficient (ADC), and pre-/post-contrast-enhanced T1-weighted imaging (T1-WI). RMs were manually segmented using volume of interest (VOI) on T2-WI, DWI/ADC, and T1-WI pre-/post-contrast imaging (1-min, 3-min post-injection) by two independent observers using two radiomics software packages for inter-package and inter-observer assessments of shape/histogram/texture features common to both packages (104 features; n = 26 patients). Intra-class correlation coefficients (ICCs) were calculated to assess inter-observer and inter-package reproducibility of radiomics measurements [good (ICC ≥ 0.8)/moderate (ICC = 0.5-0.8)/poor (ICC < 0.5)]. ML models were employed using reproducible features (between observers and packages, ICC > 0.8) to distinguish RCC from benign RM. RESULTS: Inter-package comparisons demonstrated that radiomics features from T1-WI-post-contrast had the highest proportion of good/moderate ICCs (54.8-58.6% for T1-WI-1 min), while most features extracted from T2-WI, T1-WI-pre-contrast, and ADC exhibited poor ICCs. Inter-observer comparisons found that radiomics measurements from T1-WI pre/post-contrast and T2-WI had the greatest proportion of features with good/moderate ICCs (95.3-99.1% T1-WI-post-contrast 1-min), while ADC measurements yielded mostly poor ICCs. ML models generated an AUC of 0.71 [95% confidence interval = 0.67-0.75] for diagnosis of RCC vs. benign RM. CONCLUSION: Radiomics features extracted from T1-WI-post-contrast demonstrated greater inter-package and inter-observer reproducibility compared to ADC, with fair accuracy for distinguishing RCC from benign RM. CLINICAL RELEVANCE: Knowledge of reproducibility of MRI radiomics features obtained on renal masses will aid in future study design and may enhance the diagnostic utility of radiomics models for renal mass characterization.

Prediction of locally advanced rectal cancer response to neoadjuvant chemoradiation therapy using volumetric multiparametric MRI-based radiomics.

PURPOSE: To assess the role of pretreatment multiparametric (mp)MRI-based radiomic features in predicting pathologic complete response (pCR) of locally advanced rectal cancer (LARC) to neoadjuvant chemoradiation therapy (nCRT). METHODS: This was a retrospective dual-center study including 98 patients (M/F 77/21, mean age 60 years) with LARC who underwent pretreatment mpMRI followed by nCRT and total mesorectal excision or watch and wait. Fifty-eight patients from institution 1 constituted the training set and 40 from institution 2 the validation set. Manual segmentation using volumes of interest was performed on T1WI pre-/post-contrast, T2WI and diffusion-weighted imaging (DWI) sequences. Demographic information and serum carcinoembryonic antigen (CEA) levels were collected. Shape, 1st and 2nd order radiomic features were extracted and entered in models based on principal component analysis used to predict pCR. The best model was obtained using a k-fold cross-validation method on the training set, and AUC, sensitivity and specificity for prediction of pCR were calculated on the validation set. RESULTS: Stage distribution was T3 (n = 79) or T4 (n = 19). Overall, 16 (16.3%) patients achieved pCR. Demographics, MRI TNM stage, and CEA were not predictive of pCR (p range 0.59-0.96), while several radiomic models achieved high diagnostic performance for prediction of pCR (in the validation set), with AUCs ranging from 0.7 to 0.9, with the best model based on high b-value DWI demonstrating AUC of 0.9 [95% confidence intervals: 0.67, 1], sensitivity of 100% [100%, 100%], and specificity of 81% [66%, 96%]. CONCLUSION: Radiomic models obtained from pre-treatment MRI show good to excellent performance for the prediction of pCR in patients with LARC, superior to clinical parameters and CEA. A larger study is needed for confirmation of these results.

Discovery Viewer (DV): Web-Based Medical AI Model Development Platform and Deployment Hub.

The rapid rise of artificial intelligence (AI) in medicine in the last few years highlights the importance of developing bigger and better systems for data and model sharing. However, the presence of Protected Health Information (PHI) in medical data poses a challenge when it comes to sharing. One potential solution to mitigate the risk of PHI breaches is to exclusively share pre-trained models developed using private datasets. Despite the availability of these pre-trained networks, there remains a need for an adaptable environment to test and fine-tune specific models tailored for clinical tasks. This environment should be open for peer testing, feedback, and continuous model refinement, allowing dynamic model updates that are especially important in the medical field, where diseases and scanning techniques evolve rapidly. In this context, the Discovery Viewer (DV) platform was developed in-house at the Biomedical Engineering and Imaging Institute at Mount Sinai (BMEII) to facilitate the creation and distribution of cutting-edge medical AI models that remain accessible after their development. The all-in-one platform offers a unique environment for non-AI experts to learn, develop, and share their own deep learning (DL) concepts. This paper presents various use cases of the platform, with its primary goal being to demonstrate how DV holds the potential to empower individuals without expertise in AI to create high-performing DL models. We tasked three non-AI experts to develop different musculoskeletal AI projects that encompassed segmentation, regression, and classification tasks. In each project, 80% of the samples were provided with a subset of these samples annotated to aid the volunteers in understanding the expected annotation task. Subsequently, they were responsible for annotating the remaining samples and training their models through the platform's "Training Module". The resulting models were then tested on the separate 20% hold-off dataset to assess their performance. The classification model achieved an accuracy of 0.94, a sensitivity of 0.92, and a specificity of 1. The regression model yielded a mean absolute error of 14.27 pixels. And the segmentation model attained a Dice Score of 0.93, with a sensitivity of 0.9 and a specificity of 0.99. This initiative seeks to broaden the community of medical AI model developers and democratize the access of this technology to all stakeholders. The ultimate goal is to facilitate the transition of medical AI models from research to clinical settings.

Deep Learning for Automated Measurement of Patellofemoral Anatomic Landmarks.

BACKGROUND: Patellofemoral anatomy has not been well characterized. Applying deep learning to automatically measure knee anatomy can provide a better understanding of anatomy, which can be a key factor in improving outcomes. METHODS: 483 total patients with knee CT imaging (April 2017-May 2022) from 6 centers were selected from a cohort scheduled for knee arthroplasty and a cohort with healthy knee anatomy. A total of 7 patellofemoral landmarks were annotated on 14,652 images and approved by a senior musculoskeletal radiologist. A two-stage deep learning model was trained to predict landmark coordinates using a modified ResNet50 architecture initialized with self-supervised learning pretrained weights on RadImageNet. Landmark predictions were evaluated with mean absolute error, and derived patellofemoral measurements were analyzed with Bland-Altman plots. Statistical significance of measurements was assessed by paired t-tests. RESULTS: Mean absolute error between predicted and ground truth landmark coordinates was 0.20/0.26 cm in the healthy/arthroplasty cohort. Four knee parameters were calculated, including transepicondylar axis length, transepicondylar-posterior femur axis angle, trochlear medial asymmetry, and sulcus angle. There were no statistically significant parameter differences (p > 0.05) between predicted and ground truth measurements in both cohorts, except for the healthy cohort sulcus angle. CONCLUSION: Our model accurately identifies key trochlear landmarks with ~0.20-0.26 cm accuracy and produces human-comparable measurements on both healthy and pathological knees. This work represents the first deep learning regression model for automated patellofemoral annotation trained on both physiologic and pathologic CT imaging at this scale. This novel model can enhance our ability to analyze the anatomy of the patellofemoral compartment at scale.

Early Prediction of Response of Hepatocellular Carcinoma to Yttrium-90 Radiation Segmentectomy Using a Machine Learning MR Imaging Radiomic Approach.

PURPOSE: To assess the accuracy of a machine learning (ML) approach based on magnetic resonance (MR) imaging radiomic quantification obtained before treatment and early after treatment for prediction of early hepatocellular carcinoma (HCC) response to yttrium-90 transarterial radioembolization (TARE). MATERIALS AND METHODS: In this retrospective single-center study of 76 patients with HCC, baseline and early (1-2 months) post-TARE MR images were collected. Semiautomated tumor segmentation facilitated extraction of shape, first-order histogram, and custom signal intensity-based radiomic features, which were then trained (n = 46) using a ML XGBoost model and validated on a separate cohort (n = 30) not used in training to predict treatment response assessed at 4-6 months (based on modified Response and Evaluation Criteria in Solid Tumors criteria). Performance of this ML radiomic model was compared with those of models comprising clinical parameters and standard imaging characteristics using area under the receiver operating curve (AUROC) analysis for prediction of complete response (CR). RESULTS: Seventy-six tumors with a mean (±SD) diameter of 2.6 cm ± 1.6 were included. Sixty, 12, 1, and 3 patients were classified as having CR, partial response, stable disease, and progressive disease, respectively, at 4-6 months posttreatment on the basis of MR images. In the validation cohort, the radiomic model showed good performance (AUROC, 0.89) for prediction of CR, compared with models comprising clinical and standard imaging criteria (AUROC, 0.58 and 0.59, respectively). Baseline imaging features appeared to be more heavily weighted in the radiomic model. CONCLUSIONS: The use of ML modeling of radiomic data combining baseline and early follow-up MR imaging could predict HCC response to TARE. These models need to be investigated further in an independent cohort.

Performance of spiral UTE-MRI of the lung in post-COVID patients.

Patients recovered from COVID-19 may develop long-COVID symptoms in the lung. For this patient population (post-COVID patients), they may benefit from longitudinal, radiation-free lung MRI exams for monitoring lung lesion development and progression. The purpose of this study was to investigate the performance of a spiral ultrashort echo time MRI sequence (Spiral-VIBE-UTE) in a cohort of post-COVID patients in comparison with CT and to compare image quality obtained using different spiral MRI acquisition protocols. Lung MRI was performed in 36 post-COVID patients with different acquisition protocols, including different spiral sampling reordering schemes (line in partition or partition in line) and different breath-hold positions (inspiration or expiration). Three experienced chest radiologists independently scored all the MR images for different pulmonary structures. Lung MR images from spiral acquisition protocol that received the highest image quality scores were also compared against corresponding CT images in 27 patients for evaluating diagnostic image quality and lesion identification. Spiral-VIBE-UTE MRI acquired with the line in partition reordering scheme in an inspiratory breath-holding position achieved the highest image quality scores (score range = 2.17-3.69) compared to others (score range = 1.7-3.29). Compared to corresponding chest CT images, three readers found that 81.5% (22 out of 27), 81.5% (22 out of 27) and 37% (10 out of 27) of the MR images were useful, respectively. Meanwhile, they all agreed that MRI could identify significant lesions in the lungs. The Spiral-VIBE-UTE sequence allows for fast imaging of the lung in a single breath hold. It could be a valuable tool for lung imaging without radiation and could provide great value for managing different lung diseases including assessment of post-COVID lesions.

Continuous Monitoring Versus Intermittent Auscultation of Wheezes in Patients Presenting With Acute Respiratory Distress.

BACKGROUND: Auscultation for an extended period of time using a wearable stethoscope enables objective computerized analysis and longitudinal assessment of lung sounds. However, this auscultation method differs from bedside auscultation in that clinicians are not present to optimize the quality of auscultation. No prior studies have compared these two auscultation methods. OBJECTIVE: The aim of this study was to compare intermittent auscultation using a conventional stethoscope with continuous auscultation using a wearable stethoscope for wheeze detection in patients who present with acute respiratory distress. METHODS: Patients presenting to the emergency department with acute respiratory distress were enrolled. The Strados Remote Electronic Stethoscope Platform (RESP™) was used for continuous auscultation, and intermittent auscultation was performed using a U.S. Food and Drug Administration-cleared electronic stethoscope. A recording was made with an electronic stethoscope. Subsequently, continuous recording was made using RESP™, which continued until the patient was admitted or discharged from the emergency department. The number of captured wheezes in each recording was counted and validated by two board-certified physicians. RESULTS: From May 2018 to May 2019, 43 patients were enrolled in the study. Three patients were excluded from analysis due to incomplete audio recording data. The mean length of recording was 62.3 min for continuous auscultation and 0.7 min for intermittent auscultation; 77.5% (31 of 40) of intermittent recordings contained wheezes, in contrast to 85% (34 of 40) of continuous recordings. CONCLUSIONS: Extending the duration of auscultation using a wearable stethoscope in a noisy clinical environment showed comparable performance to standard of care intermittent auscultation in identifying patients who have wheezes.