A5044687271- Radiomics and Machine Learning in Medical Imaging
- Artificial Intelligence in Healthcare and Education
- Advanced X-ray and CT Imaging
- Orthopaedic implants and arthroplasty
- Total Knee Arthroplasty Outcomes
- AI in cancer detection
- Orthopedic Infections and Treatments
- COVID-19 diagnosis using AI
- Medical Imaging and Analysis
- Radiology practices and education
- Radiation Dose and Imaging
- Machine Learning in Healthcare
- Esophageal Cancer Research and Treatment
- COVID-19 Clinical Research Studies
- Image and Signal Denoising Methods
- Dental Radiography and Imaging
- Medical Imaging Techniques and Applications
- Hip and Femur Fractures
- Colorectal Cancer Screening and Detection
- Advanced Image Processing Techniques
- Hip disorders and treatments
- Spinal Fractures and Fixation Techniques
- Topic Modeling
- Ultrasound Imaging and Elastography
- Explainable Artificial Intelligence (XAI)
Mayo Clinic in Arizona
2021-2025
Mayo Clinic
2020-2025
Mayo Clinic in Florida
2022-2025
University of California, Los Angeles
2024-2025
University of Washington
2025
University of British Columbia
2025
Trinity Health
2024
Wayne State University
2024
University of Colorado Anschutz Medical Campus
2024
Ottawa Hospital
2024
Minimizing bias is critical to adoption and implementation of machine learning (ML) in clinical practice. Systematic mathematical biases produce consistent reproducible differences between the observed expected performance ML systems, resulting suboptimal performance. Such can be traced back various phases development: data handling, model development, evaluation. This report presents 12 practices during handling an study, explains how those lead biases, describes what may done mitigate...
There is a growing demand for high-resolution (HR) medical images both clinical and research applications. Image quality inevitably traded off with acquisition time, which in turn impacts patient comfort, examination costs, dose, motion-induced artifacts. For many image-based tasks, increasing the apparent spatial resolution perpendicular plane to produce multi-planar reformats or 3D commonly used. Single-image super-resolution (SR) promising technique provide HR based on deep learning...
There are increasing concerns about the bias and fairness of artificial intelligence (AI) models as they put into clinical practice. Among steps for implementing machine learning tools workflow, model development is an important stage where different types biases can occur. This report focuses on four aspects such may arise: data augmentation, loss function, optimizers, transfer learning. emphasizes appropriate considerations practices that mitigate in radiology AI studies.
The increasing use of machine learning (ML) algorithms in clinical settings raises concerns about bias ML models. Bias can arise at any step creation, including data handling, model development, and performance evaluation. Potential biases the be minimized by implementing these steps correctly. This report focuses on evaluation discusses fitness, as well a set toolboxes: namely, metrics, interpretation maps, uncertainty quantification. By discussing strengths limitations each toolbox, our...
Background Multiparametric MRI can help identify clinically significant prostate cancer (csPCa) (Gleason score ≥7) but is limited by reader experience and interobserver variability. In contrast, deep learning (DL) produces deterministic outputs. Purpose To develop a DL model to predict the presence of csPCa using patient-level labels without information about tumor location compare its performance with that radiologists. Materials Methods Data from patients known who underwent January 2017...
Background: Establishing imaging registries for large patient cohorts is challenging because manual labeling tedious and relying solely on DICOM (digital communications in medicine) metadata can result errors. We endeavored to establish an automated hip pelvic radiography registry of total arthroplasty (THA) patients by utilizing deep-learning pipelines. The aims the study were (1) utilize these pipelines identify all radiographs with appropriate annotation laterality presence or absence...
To develop a deep learning model that segments intracranial structures on head CT scans.In this retrospective study, primary dataset containing 62 normal noncontrast scans from patients (mean age, 73 years; age range, 27-95 years) acquired between August and December 2018 was used for development. Eleven were manually annotated the axial oblique series. The split into 40 training, 10 validation, 12 testing. After initial eight configurations evaluated validation highest performing test...
To develop a multimodal machine learning-based pipeline to predict patient-specific risk of dislocation following primary total hip arthroplasty (THA).This study retrospectively evaluated 17 073 patients who underwent THA between 1998 and 2018. A test set 1718 was held out. hybrid network EfficientNet-B4 Swin-B transformer developed classify according 5-year outcomes from preoperative anteroposterior pelvic radiographs clinical characteristics (demographics, comorbidities, surgical...
Abstract Background Defining standards is the first step toward quality assurance and improvement of educational programs. This study aimed at developing validating a set national for Undergraduate Medical Education (UME) program through an accreditation system in Iran using World Federation (WFME) framework. Methods The draft was prepared consultative workshops with participation different UME stakeholders. Subsequently, were sent to medical schools directors asked complete web-based...
Radiographic markers contain protected health information that must be removed before public release. This work presents a deep learning algorithm localizes radiographic and selectively removes them to enable de-identified data sharing. The authors annotated 2000 hip pelvic radiographs train an object detection computer vision model. Data were split into training, validation, test sets at the patient level. Extracted then characterized using image processing algorithm, potentially useful...
The application of deep learning (DL) in medicine introduces transformative tools with the potential to enhance prognosis, diagnosis, and treatment planning. However, ensuring transparent documentation is essential for researchers reproducibility refine techniques. Our study addresses unique challenges presented by DL medical imaging developing a comprehensive checklist using Delphi method reliability this dynamic field. We compiled preliminary based on review existing checklists relevant...
Dislocation is the most common reason for early revision following total hip arthroplasty (THA). More than 40 years ago, Lewinnek et al. proposed an acetabular "safe zone" to avoid dislocation. While novel at time, their study was substantially limited according modern standards. The purpose of this determine optimal cup positioning during THA as well effect surgical approach on topography safe zone and hazard dislocation.Primary THAs that had been performed a single institution from 2000...
Systematic literature reviews and meta-analyses are essential for synthesizing research insights, but they remain time-intensive labor-intensive due to the iterative processes of screening, evaluation, data extraction. This paper introduces evaluates LatteReview, a Python-based framework that leverages large language models (LLMs) multi-agent systems automate key elements systematic review process. Designed streamline workflows while maintaining rigor, LatteReview utilizes modular agents...