Thai Hospital's AI Surpasses 500,000 X-Rays
What happened
Siriraj Hospital in Thailand reports its artificial intelligence for reading chest X-rays has been deployed in over 500,000 cases with an accuracy rate exceeding 95%. The system has been rolled out to more than 145 public hospitals, showcasing a large-scale, national adoption of AI in diagnostic workflows.
Why it matters
- The AI, co-developed with the company Perceptra and named "Inspectra," was trained on a dataset of half a million chest X-rays and runs on NVIDIA DGX A100 systems. - Siriraj Hospital is extending the AI's application beyond chest X-rays, with plans to develop similar tools for interpreting brain CT scans and screening for breast cancer with digital mammography. - This type of AI adoption addresses a critical global radiologist shortage, which in the U.S. is projected to reach a deficit of up to 42,000 radiologists by 2033. - The efficiency gains from such AI support the broader shift of imaging services to non-hospital settings; currently, about 40% of all radiology volume in the U.S. is performed in outpatient imaging centers or clinics. - While hundreds of radiology AI tools have received FDA clearance, direct reimbursement remains a hurdle; most AI algorithms are associated with temporary Category III CPT codes used for tracking rather than payment, making the return on investment primarily about operational efficiency. - In the U.S. market, many FDA-cleared chest X-ray AI tools are focused on computer-assisted triage and notification for emergent conditions like pneumothorax and pleural effusion, immediately flagging urgent cases for radiologists. - The growth of AI-driven tools aligns with the expansion of the mobile medical imaging market, which is forecast to grow from $17.31 billion in 2026 to $21.13 billion by 2031, fueled by the decentralization of care and favorable reimbursement for remote diagnostics. - The implementation of AI is a key strategy for mitigating radiologist burnout and increasing throughput as imaging demand rises; in emergency departments at some U.S. trauma centers, imaging volumes have increased by over 34%.
Key numbers
- Siriraj Hospital in Thailand reports its artificial intelligence for reading chest X-rays has been deployed in over 500,000 cases with an accuracy rate exceeding 95%.
- The system has been rolled out to more than 145 public hospitals, showcasing a large-scale, national adoption of AI in diagnostic workflows.
- - The AI, co-developed with the company Perceptra and named "Inspectra," was trained on a dataset of half a million chest X-rays and runs on NVIDIA DGX A100 systems.
- is projected to reach a deficit of up to 42,000 radiologists by 2033.
What happens next
- Siriraj Hospital is extending the AI's application beyond chest X-rays, with plans to develop similar tools for interpreting brain CT scans and screening for breast cancer with digital mammography.
Quick answers
What happened in Thai Hospital's AI Surpasses 500,000 X-Rays?
Siriraj Hospital in Thailand reports its artificial intelligence for reading chest X-rays has been deployed in over 500,000 cases with an accuracy rate exceeding 95%. The system has been rolled out to more than 145 public hospitals, showcasing a large-scale, national adoption of AI in diagnostic workflows.
Why does Thai Hospital's AI Surpasses 500,000 X-Rays matter?
The AI, co-developed with the company Perceptra and named "Inspectra," was trained on a dataset of half a million chest X-rays and runs on NVIDIA DGX A100 systems. Siriraj Hospital is extending the AI's application beyond chest X-rays, with plans to develop similar tools for interpreting brain CT scans and screening for breast cancer with digital mammography. This type of AI adoption addresses a critical global radiologist shortage, which in the U.S. is projected to reach a deficit of up to 42,000 radiologists by 2033. The efficiency gains from such AI support the broader shift of imaging services to non-hospital settings; currently, about 40% of all radiology volume in the U.S. is performed in outpatient imaging centers or clinics. While hundreds of radiology AI tools have received FDA clearance, direct reimbursement remains a hurdle; most AI algorithms are associated with temporary Category III CPT codes used for tracking rather than payment, making the return on investment primarily about operational efficiency. In the U.S. market, many FDA-cleared chest X-ray AI tools are focused on computer-assisted triage and notification for emergent conditions like pneumothorax and pleural effusion, immediately flagging urgent cases for radiologists. The growth of AI-driven tools aligns with the expansion of the mobile medical imaging market, which is forecast to grow from $17.31 billion in 2026 to $21.13 billion by 2031, fueled by the decentralization of care and favorable reimbursement for remote diagnostics. The implementation of AI is a key strategy for mitigating radiologist burnout and increasing throughput as imaging demand rises; in emergency departments at some U.S. trauma centers, imaging volumes have increased by over 34%.
