Search

We aimed to develop deep learning models using longitudinal chest X-rays (CXRs) and clinical data to predict in-hospital mortality of COVID-19 patients in the intensive care unit (ICU).Six hundred fifty-four patients (212 deceased, 442 alive, 5645 total CXRs) were identified across two institutions. Imaging and clinical data from one institution were used to train five longitudinal transformer-based networks applying five-fold cross-validation. The models were tested on data from the other institution, and pairwise comparisons were used to determine the best-performing models.A higher proportion of deceased patients had elevated white blood cell count, decreased absolute lymphocyte count, elevated creatine concentration, and incidence of cardiovascular and chronic kidney disease. A model based on pre-ICU CXRs achieved an AUC of 0.632 and an accuracy of 0.593, and a model based on ICU CXRs achieved an AUC of 0.697 and an accuracy of 0.657. A model based on all longitudinal CXRs (both pre-ICU and ICU) achieved an AUC of 0.702 and an accuracy of 0.694. A model based on clinical data alone achieved an AUC of 0.653 and an accuracy of 0.657. The addition of longitudinal imaging to clinical data in a combined model significantly improved performance, reaching an AUC of 0.727 (p = 0.039) and an accuracy of 0.732.The addition of longitudinal CXRs to clinical data significantly improves mortality prediction with deep learning for COVID-19 patients in the ICU.• Deep learning was used to predict mortality in COVID-19 ICU patients. • Serial radiographs and clinical data were used. • The models could inform clinical decision-making and resource allocation.

In the era of data-driven medicine, rapid access and accurate interpretation of medical images are becoming increasingly important. The DICOM Image ANalysis and Archive (DIANA) system is an open-source, lightweight, and scalable Python interface that enables users to interact with hospital Picture Archiving and Communications Systems (PACS) to access such data. In this work, DIANA functionality was detailed and evaluated in the context of retrospective PACS data retrieval and two prospective clinical artificial intelligence (AI) pipelines: bone age (BA) estimation and intra-cranial hemorrhage (ICH) detection. DIANA orchestrates activity beginning with post-acquisition study discovery and ending with online notifications of findings. For AI applications, system latency (exam completion to system report time) was quantified and compared to that of clinicians (exam completion to initial report creation time). Mean DIANA latency was 9.04 ± 3.83 and 20.17 ± 10.16 min compared to clinician latency of 51.52 ± 58.9 and 65.62 ± 110.39 min for BA and ICH, respectively, with DIANA latencies being significantly lower (p

BackgroundClinical implementation of polygenic risk scores (PRS) for precision prevention depends on the utility and barriers primary care physicians (PCPs) perceive to their use.MethodsAn online survey asked PCPs in a national database about the clinical utility of PRS they perceived for categories of medical decision-making and perceived benefits of and barriers to that use. Latent class analysis (LCA) was used to identify subgroups of PCPs based on response patterns.ResultsAmong 367 respondents (email open rate 10.8%; participation rate 96.3%; completion rate 93.1%), mean (SD) age was 54.9 (12.9) years, 137 (37.3%) were female, and mean (SD) time since medical school graduation was 27.2 (13.3) years. Respondents reported greater perceived utility for more clinical action (e.g., earlier or more intensive screening, preventive medications, or lifestyle modification) for patients with high-risk PRS than for delayed or discontinued prevention actions for low-risk patients (pSkeptics (n=83, 22.6%) endorsed less agreement with individual clinical utilities, saw patient anxiety and insurance discrimination as significant barriers, and agreed less often that PRS could help patients make better health decisions. Learners (n=134, 36.5%) and enthusiasts (n=150, 40.9%) expressed similar levels of agreement that PRS had utility for preventive actions and that PRS could be useful for patient decision-making. Compared with enthusiasts, however, learners perceived greater barriers to the clinical use of PRS.ConclusionPCPs generally endorsed using PRS to guide medical decision-making about preventive care, with a preference for more clinical action over less. Barriers identified suggest interventions to address the needs and concerns of PCPs along the spectrum of acceptance and uptake.

In this study, we developed a deep learning pipeline that detects large vessel occlusion (LVO) and predicts functional outcome based on computed tomography angiography (CTA) images to improve the management of the LVO patients.A series identifier picked out 8650 LVO-protocoled studies from 2015 to 2019 at Rhode Island Hospital with an identified thin axial series that served as the data pool. Data were annotated into 2 classes: 1021 LVOs and 7629 normal. The Inception-V1 I3D architecture was applied for LVO detection. For outcome prediction, 323 patients undergoing thrombectomy were selected. A 3D convolution neural network (CNN) was used for outcome prediction (30-day mRS) with CTA volumes and embedded pre-treatment variables as inputs.For LVO-detection model, CTAs from 8,650 patients (median age 68 years, interquartile range (IQR): 58-81; 3934 females) were analyzed. The cross-validated AUC for LVO vs. not was 0.74 (95% CI: 0.72-0.75). For the mRS classification model, CTAs from 323 patients (median age 75 years, IQR: 63-84; 164 females) were analyzed. The algorithm achieved a test AUC of 0.82 (95% CI: 0.79-0.84), sensitivity of 89%, and specificity 66%. The two models were then integrated with hospital infrastructure where CTA was collected in real-time and processed by the model. If LVO was detected, interventionists were notified and provided with predicted clinical outcome information.3D CNNs based on CTA were effective in selecting LVO and predicting LVO mechanical thrombectomy short-term prognosis. End-to-end AI platform allows users to receive immediate prognosis prediction and facilitates clinical workflow.

There currently lacks a noninvasive and accurate method to distinguish benign and malignant ovarian lesion prior to treatment. This study developed a deep learning algorithm that distinguishes benign from malignant ovarian lesion by applying a convolutional neural network on routine MR imaging. Five hundred forty-five lesions (379 benign and 166 malignant) from 451 patients from a single institution were divided into training, validation, and testing set in a 7:2:1 ratio. Model performance was compared with four junior and three senior radiologists on the test set. Compared with junior radiologists averaged, the final ensemble model combining MR imaging and clinical variables had a higher test accuracy (0.87 vs 0.64, p

e13551 Background: Machine learning (ML) algorithms to predict cancer survival have recently been reported for a number of sarcoma subtypes, but none have investigated undifferentiated pleomorphic sarcoma (UPS). ML is a powerful tool that has the potential to better prognosticate UPS. Methods: The Surveillance, Epidemiology, and End Results (SEER) database was queried from 2004 to 2015 for cases of histologically confirmed undifferentiated pleomorphic sarcoma (UPS) and malignant fibrous histiocytoma (MFH). Patient, tumor, and treatment characteristics were recorded, and various machine learning (ML) models were built to predict 1-, 3-, and 5-year survival. The best performing ML models were externally validated using an institutional cohort of UPS patients. Results: All ML models performed best at the 1-year time point and worst at the 5-year time point. On internal validation within the SEER cohort, the best models had c-statistics of 0.71 to 0.73 at the 5-year time point. Similarly, all ML models performed best at 1-year and worst at 5-year on external validation. The best performing models had c-statistics of 0.81 at the 5-year time point on external validation, demonstrating good performance in survival prediction. Conclusions: Machine learning models perform well for survival prediction in UPS, though this sarcoma subtype may be more difficult to prognosticate than other subtypes. Future studies are needed to further validate the machine learning approach for UPS prognostication. [Table: see text]

PURPOSE: To assess the imaging performance and radiation dose characteristics of the O-arm CBCT imaging system (Medtronic Inc., Littleton MA) and demonstrate the potential for improved image quality and reduced dose via model-based image reconstruction (MBIR). METHODS: Two main studies were performed to investigate previously unreported characteristics of the O-arm system. First is an investigation of dose and 3D image quality achieved with filtered back-projection (FBP) — including enhancements in geometric calibration, handling of lateral truncation and detector saturation, and incorporation of an isotropic apodization filter. Second is implementation of an MBIR algorithm based on Huber-penalized likelihood estimation (PLH) and investigation of image quality improvement at reduced dose. Each study involved measurements in quantitative phantoms as a basis for analysis of contrast-to-noise ratio and spatial resolution as well as imaging of a human cadaver to test the findings under realistic imaging conditions. RESULTS: View-dependent calibration of system geometry improved the accuracy of reconstruction as quantified by the full-width at half maximum of the point-spread function — from 0.80 to 0.65 mm — and yielded subtle but perceptible improvement in high-contrast detail of bone (e.g., temporal bone). Standard technique protocols for the head and body imparted absorbed dose of 16 and 18 mGy, respectively. For low-to-medium contrast (

Background: To develop a deep learning model to classify primary bone tumors from preoperative radiographs and compare performance with radiologists. Methods: A total of 1356 patients (2899 images) with histologically confirmed primary bone tumors and pre-operative radiographs were identified from five institutions’ pathology databases. Manual cropping was performed by radiologists to label the lesions. Binary discriminatory capacity (benign versus not-benign and malignant versus not-malignant) and three-way classification (benign versus intermediate versus malignant) performance of our model were evaluated. The generalizability of our model was investigated on data from external test set. Final model performance was compared with interpretation from five radiologists of varying level of experience using the Permutations tests. Findings: For benign vs. not benign, model achieved area under curve (AUC) of 0•894 and 0•877 on cross-validation and external testing, respectively. For malignant vs. not malignant, model achieved AUC of 0•907 and 0•916 on cross-validation and external testing, respectively. For three-way classification, model achieved 72•1% accuracy vs. 74•6% and 72•1% for the two subspecialists on cross-validation (p = 0•03 and p = 0•52, respectively). On external testing, model achieved 73•4% accuracy vs. 69•3%, 73•4%, 73•1%, 67•9%, and 63•4% for the two subspecialists and three junior radiologists (p = 0•14, p = 0•89, p = 0•93, p = 0•02, p

Purpose To compare overall survival (OS) of ablation with no treatment for patients with advanced stage non-small cell lung cancer. Methods Patients with clinical stage IIIB (T1–4N3M0, T4N2M0) and stage IV (T1–4N0–3M1) non-small cell lung cancer, in accordance with the American Joint Committee on Cancer, 7th edition, who did not receive treatment or who received ablation as their sole primary treatment besides chemotherapy from 2004 to 2014, were identified from the National Cancer Data Base. OS was estimated using the Kaplan-Meier method and evaluated by log-rank test, univariate and multivariate Cox proportional hazard regression, and propensity score-matched analysis. Relative survival analyses comparing age- and sex-matched United States populations were performed. Results A total of 140,819 patients were included. The 1-, 2-, 3- and 5-year survival rates relative to age- and sex-matched United States population were 28%, 18%, 12%, and 10%, respectively, for ablation (n = 249); and 30%, 15%, 9%, and 5%, respectively for no treatment (n = 140,570). Propensity score matching resulted in 249 patients in the ablation group versus 498 patients in the no-treatment group. After matching, ablation was associated with longer OS than that in the no-treatment group (median, 5.9 vs 4.7 months, respectively; hazard ratio, 0.844; 95% confidence interval, 0.719–0.990; P = .037). These results persisted in patients with an initial tumor size of ≤3 cm. Conclusions Preliminary results suggest ablation may be associated with longer OS in patients with late-stage non-small cell lung cancer than survival in those who received no treatment.

A method for x-ray-guided robotic positioning of surgical instruments is reported and evaluated in preclinical studies of spine pedicle screw placement with the aim of improving delivery of transpedicle drills and screws. The known-component registration (KC-Reg) algorithm was used to register the 3D patient CT and the surface model of a drill guide to intraoperatively acquired 2D radiographs. Resulting transformations, combined with offline hand-eye calibration, drive a robotically-held drill guide to target trajectories established in the preoperative patient CT. The proposed method was assessed against more conventional surgical tracker guidance, and robustness to clinically realistic errors was tested in phantom and cadaver studies. Target registration error (TRE) was computed as drill guide deviation from the planned trajectory. The KC-Reg approach resulted in 1.51 ± 0.51 mm error at tooltip and 1.01 ± 0.92° in approach angle, showing comparable performance to the tracker-guided approach. In cadaver studies with anatomical deformation, TRE of 2.31 ± 1.05 mm and 0.66 ± 0.62° were observed, with statistically improved performance over a surgical tracker through registration of locally rigid bony anatomy. X-ray guidance offers an accurate means of driving robotic systems that is compatible with conventional fluoroscopic workflow. Specifically, such procedures involve multi-planar fluoroscopic views that are qualitatively interpreted by the surgeon; the KC-Reg approach accomplishes this using the same multi-planar views to provide greater quantitative accuracy and valuable guidance and QA. The method was robust against anatomical deformation due to the radiographic scene's local nature used in registration, presenting a potentially major surgical benefit.

PURPOSE: To produce and maintain a database of National Institutes of Health (NIH) funding of the American Association of Physicists in Medicine (AAPM) members, to perform a top‐level analysis of these data, and to make these data (hereafter referred to as the AAPM research database) available for the use of the AAPM and its members. METHODS: NIH‐funded research dating back to 1985 is available for public download through the NIH exporter website, and AAPM membership information dating back to 2002 was supplied by the AAPM. To link these two sources of data, a data mining algorithm was developed in Matlab. The false‐positive rate was manually estimated based on a random sample of 100 records, and the false‐negative rate was assessed by comparing against 99 member‐supplied PI _ ID numbers. The AAPM research database was queried to produce an analysis of trends and demographics in research funding dating from 2002 to 2015. RESULTS: A total of 566 PI _ ID numbers were matched to AAPM members. False‐positive and ‐negative rates were respectively 4% (95% CI: 1–10%, N = 100) and 10% (95% CI: 5–18%, N = 99). Based on analysis of the AAPM research database, in 2015 the NIH awarded $USD 110M to members of the AAPM. The four NIH institutes which historically awarded the most funding to AAPM members were the National Cancer Institute, National Institute of Biomedical Imaging and Bioengineering, National Heart Lung and Blood Institute, and National Institute of Neurological Disorders and Stroke. In 2015, over 85% of the total NIH research funding awarded to AAPM members was via these institutes, representing 1.1% of their combined budget. In the same year, 2.0% of AAPM members received NIH funding for a total of $116M, which is lower than the historic mean of $120M (in 2015 USD). CONCLUSIONS: A database of NIH‐funded research awarded to AAPM members has been developed and tested using a data mining approach, and a top‐level analysis of funding trends has been performed. Current funding of AAPM members is lower than the historic mean. The database will be maintained by members of the Working group for the development of a research database (WGDRD) on an annual basis, and is available to the AAPM, its committees, working groups, and members for download through the AAPM electronic content website. A wide range of questions regarding financial and demographic funding trends can be addressed by these data. This report has been approved for publication by the AAPM Science Council.

STUDY DESIGN An automatic radiographic labeling algorithm called "LevelCheck" was analyzed as a means of decision support for target localization in spine surgery. The potential clinical utility and scenarios in which LevelCheck is likely to be the most beneficial were assessed in a retrospective clinical data set (398 cases) in terms of expert consensus from a multi-reader study (three spine surgeons). OBJECTIVE The aim of this study was to evaluate the potential utility of the LevelCheck algorithm for vertebrae localization. SUMMARY OF BACKGROUND DATA Three hundred ninety-eight intraoperative radiographs and 178 preoperative computed tomographic (CT) images for patients undergoing spine surgery in cervical, thoracic, lumbar regions. METHODS Vertebral labels annotated in preoperative CT image were overlaid on intraoperative radiographs via 3D-2D registration. Three spine surgeons assessed the radiographs and LevelCheck labeling according to a questionnaire evaluating performance, utility, and suitability to surgical workflow. Geometric accuracy and registration run time were measured for each case. RESULTS LevelCheck was judged to be helpful in 42.2% of the cases (168/398), to improve confidence in 30.6% of the cases (122/398), and in no case diminished performance (0/398), supporting its potential as an independent check and assistant to decision support in spine surgery. The clinical contexts for which the method was judged most likely to be beneficial included the following scenarios: images with a lack of conspicuous anatomical landmarks; level counting across long spine segments; vertebrae obscured by other anatomy (e.g., shoulders); poor radiographic image quality; and anatomical variations/abnormalities. The method demonstrated 100% geometric accuracy (i.e., overlaid labels within the correct vertebral level in all cases) and did not introduce ambiguity in image interpretation. CONCLUSION LevelCheck is a potentially useful means of decision support in vertebral level localization in spine surgery. LEVEL OF EVIDENCE N/A.

Intraoperative cone-beam CT (CBCT) is increasingly used for surgical navigation and validation of device placement. In spinal deformity correction, CBCT provides visualization of pedicle screws and fixation rods in relation to adjacent anatomy. This work reports and evaluates a method that uses prior information regarding such surgical instrumentation for improved metal artifact reduction (MAR). The known-component MAR (KC-MAR) approach achieves precise localization of instrumentation in projection images using rigid or deformable 3D-2D registration of component models, thereby overcoming residual errors associated with segmentation-based methods. Projection data containing metal components are processed via 2D inpainting of the detector signal, followed by 3D filtered back-projection (FBP). Phantom studies were performed to identify nominal algorithm parameters and quantitatively investigate performance over a range of component material composition and size. A cadaver study emulating screw and rod placement in spinal deformity correction was conducted to evaluate performance under realistic clinical imaging conditions. KC-MAR demonstrated reduction in artifacts (standard deviation in voxel values) across a range of component types and dose levels, reducing the artifact to 5-10 HU. Accurate component delineation was demonstrated for rigid (screw) and deformable (rod) models with sub-mm registration errors, and a single-pixel dilation of the projected components was found to compensate for partial-volume effects. Artifacts associated with spine screws and rods were reduced by 40%-80% in cadaver studies, and the resulting images demonstrated markedly improved visualization of instrumentation (e.g. screw threads) within cortical margins. The KC-MAR algorithm combines knowledge of surgical instrumentation with 3D image reconstruction in a manner that overcomes potential pitfalls of segmentation. The approach is compatible with FBP-thereby maintaining simplicity in a manner that is consistent with surgical workflow-or more sophisticated model-based reconstruction methods that could further improve image quality and/or help reduce radiation dose.

A method for x-ray image-guided robotic instrument positioning is reported and evaluated in preclinical studies of spinal pedicle screw placement with the aim of improving delivery of transpedicle K-wires and screws. The known-component (KC) registration algorithm was used to register the three-dimensional patient CT and drill guide surface model to intraoperative two-dimensional radiographs. Resulting transformations, combined with offline hand-eye calibration, drive the robotically held drill guide to target trajectories defined in the preoperative CT. The method was assessed in comparison with a more conventional tracker-based approach, and robustness to clinically realistic errors was tested in phantom and cadaver. Deviations from planned trajectories were analyzed in terms of target registration error (TRE) at the tooltip (mm) and approach angle (deg). In phantom studies, the KC approach resulted in [Formula: see text] and [Formula: see text], comparable with accuracy in tracker-based approach. In cadaver studies with realistic anatomical deformation, the KC approach yielded [Formula: see text] and [Formula: see text], with statistically significant improvement versus tracker ([Formula: see text] and [Formula: see text]). Robustness to deformation is attributed to relatively local rigidity of anatomy in radiographic views. X-ray guidance offered accurate robotic positioning and could fit naturally within clinical workflow of fluoroscopically guided procedures.