Medical AI for Chest Radiography
Precision in every image.
Insight in every reading.
TORAH Vision AI classifies 14 abnormal findings on high-resolution chest X-rays and visualizes the regions the AI attends to, supporting clinicians as they read.
- ISO 13485 Quality Management System
- Digital Medical Device GMP
- CES 2026 Innovation Awards Honoree
01Solution
TORAH Vision AI
From image upload to review. Finding classification, lesion localization and Grad-CAM visualization, delivered in a single analysis report.
-
01
Multi-finding analysis
Classifies 14 abnormal findings and No Finding in a single pass.
-
02
Visual explanation
Bounding boxes and Grad-CAM heatmaps show where the AI focused, overlaid on the original image.
-
03
Integrated report
Per-finding scores and visualizations are combined into one report.
- Learn more
02Analysis Findings
14 abnormal chest findings
Covering the lung parenchyma, pleura, heart and great vessels — plus a separate No Finding class for normal studies.
- 01Aortic Enlargement
- 02Atelectasis
- 03Calcification
- 04Cardiomegaly
- 05Consolidation
- 06ILDInterstitial Lung Disease
- 07Infiltration
- 08Lung Opacity
- 09Nodule/MassNodule / Mass
- 10Other Lesion
- 11Pleural Effusion
- 12Pleural Thickening
- 13Pneumothorax
- 14Pulmonary Fibrosis
- Findings
- 14 + 114 abnormal findings plus No Finding
- Imaging
- DICOMNative support for high-resolution medical images
- Explainability
- Grad-CAMVisualization of AI attention regions
- Quality
- ISO 13485Medical device quality management
03Workflow
From image input to results
A five-step analysis process delivers results ready for clinical review.
-
STEP 01
Image input
Upload a chest X-ray in DICOM or another supported format.
-
STEP 02
Preprocessing
The high-resolution image is normalized and prepared for analysis.
-
STEP 03
AI analysis
The deep learning model analyzes the image and classifies findings.
-
STEP 04
Grad-CAM visualization
Regions the AI attended to are rendered as a heatmap.
-
STEP 05
Review
Review the report with per-finding scores and visualizations.
04Quality & Recognition
Built to international standards,
recognized worldwide
TORAH designs and manufactures its products under an ISO 13485 quality management system and Digital Medical Device GMP.
05News
Latest from TORAH
CES 2026 Innovation Awards Honoree
TORAH Vision AI was named a CES 2026 Innovation Awards Honoree in the Artificial Intelligence category.
Excellence Award for AI Prognosis Research
Recognized for research on AI-based prognosis prediction in bacteremia patients with pulmonary infection.
Hangzhou Global Digital Industry Expo
Exhibited and demonstrated TORAH Vision AI in Hangzhou, China.
Contact
Let's talk about
medical AI for your institution
Product adoption, demo requests, clinical collaboration or international partnerships — our team responds within one business day.