A5041963398- COVID-19 diagnosis using AI
- Tuberculosis Research and Epidemiology
- Radiomics and Machine Learning in Medical Imaging
- Functional Brain Connectivity Studies
- EEG and Brain-Computer Interfaces
- Neural dynamics and brain function
- Infectious Diseases and Tuberculosis
- Medical Image Segmentation Techniques
- Robotics and Sensor-Based Localization
- Neuroscience and Music Perception
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Sleep and Wakefulness Research
- Lung Cancer Diagnosis and Treatment
- Artificial Intelligence in Healthcare and Education
- Obsessive-Compulsive Spectrum Disorders
- Pneumonia and Respiratory Infections
- Video Analysis and Summarization
Google (United States)
2021-2024
Tel Aviv University
2009-2014
Tel Aviv Sourasky Medical Center
2012
Background The World Health Organization (WHO) recommends chest radiography to facilitate tuberculosis (TB) screening. However, radiograph interpretation expertise remains limited in many regions. Purpose To develop a deep learning system (DLS) detect active pulmonary TB on radiographs and compare its performance that of radiologists. Materials Methods A DLS was trained tested using retrospective (acquired between 1996 2020) from 10 countries. improve generalization, large-scale pretraining,...
Chest radiography (CXR) is the most widely-used thoracic clinical imaging modality and crucial for guiding management of cardiothoracic conditions. The detection specific CXR findings has been main focus several artificial intelligence (AI) systems. However, wide range possible abnormalities makes it impractical to build systems detect every condition. In this work, we developed evaluated an AI system classify CXRs as normal or abnormal. For development, used a de-identified dataset 248,445...
ORIGINAL RESEARCH article Front. Hum. Neurosci., 12 April 2012Sec. Motor Neuroscience https://doi.org/10.3389/fnhum.2012.00079
Using artificial intelligence (AI) to interpret chest X-rays (CXRs) could support accessible triage tests for active pulmonary tuberculosis (TB) in resource-constrained settings. The performance of two cloud-based CXR AI systems - one detect TB and the other abnormalities a population with high human immunodeficiency virus (HIV) burden was evaluated. We recruited 1978 adults who had symptoms, were close contacts known patients, or newly diagnosed HIV at three clinical sites. TB-detecting (TB...
Tuberculosis (TB) is a top-10 cause of death worldwide. Though the WHO recommends chest radiographs (CXRs) for TB screening, limited availability CXR interpretation barrier. We trained deep learning system (DLS) to detect active pulmonary using CXRs from 9 countries across Africa, Asia, and Europe, utilized large-scale pretraining, attention pooling, noisy student semi-supervised learning. Evaluation was on (1) combined test set spanning China, India, US, Zambia, (2) an independent mining...