New AI Tool Mimics Radiologist Gaze to Read Chest X-Rays

Artificial intelligence (AI) can scan a chest X-ray and diagnose if an abnormality is fluid in the lungs, an enlarged heart or cancer. But being right is not enough, said Ngan Le, a University of Arkansas assistant professor of computer science and computer engineering. We should understand how the computer makes its diagnosis, yet most AI systems are black boxes whose "thought process" even their creators cannot explain.

"When people understand the reasoning process and limitations behind AI decisions, they are more likely to trust and embrace the technology," Le said.

Le and her colleagues developed a transparent, and highly accurate, AI framework for reading chest X-rays called ItpCtrl-AI, which stands for interpretable and controllable artificial intelligence.

The team explained their approach in "ItpCtrl-AI: End-to-end interpretable and controllable artificial intelligence by modeling radiologists’ intentions," published in the current issue of Artificial Intelligence in Medicine.

The researchers taught the computer to look at chest X-rays like a radiologist. The gaze of radiologists, both where they looked and how long they focused on a specific area, was recorded as they reviewed chest X-rays. The heat map created from that eye-gaze dataset showed the computer where to search for abnormalities and what section of the image required less attention.

Creating an AI framework that uses a clear, transparent method to reach conclusions - in this case a gaze heat map - helps researchers adjust and correct the computer so it can provide more accurate results. In a medical context, transparency also bolsters the trust of doctors and patients in an AI-generated diagnosis.

"If an AI medical assistant system diagnoses a condition, doctors need to understand why it made that decision to ensure it is reliable and aligns with medical expertise," Le said.

A transparent AI framework is also more accountable, a legal and ethical concern in areas with high stakes, such as medicine, self-driving vehicles or financial markets. Because doctors know how ItpCtrl-AI works, they can take responsibility for its diagnosis.

"If we don't know how a system is making decisions, it’s challenging to ensure it is fair, unbiased, or aligned with societal values," Le said.

Le and her team, in collaboration with the MD Anderson Cancer Center in Houston, are now working to refine ItpCtrl-AI so it can read more complex, three-dimensional CT scans.

Trong-Thang Pham, Jacob Brecheisen, Carol C. Wu, Hien Nguyen, Zhigang Deng, Donald Adjeroh, Gianfranco Doretto, Arabinda Choudhary, Ngan Le.
ItpCtrl-AI: End-to-end interpretable and controllable artificial intelligence by modeling radiologists' intentions.
Artificial Intelligence in Medicine, 2025. doi: 10.1016/j.artmed.2024.103054

Most Popular Now

Personalized Breast Cancer Prevention No…

A new telemedicine service for personalised breast cancer prevention has launched at preventcancer.co.uk. It allows women aged 30 to 75 across the UK to understand their risk of developing breast...

New App may Help Caregivers of People Ge…

A new study by investigators from Mass General Brigham showed that a new app they created can help improve the quality of life for caregivers of patients undergoing bone marrow...

An App to Detect Heart Attacks and Strok…

A potentially lifesaving new smartphone app can help people determine if they are suffering heart attacks or strokes and should seek medical attention, a clinical study suggests. The ECHAS app (Emergency...

A Machine Learning Tool for Diagnosing, …

Scientists aiming to advance cancer diagnostics have developed a machine learning tool that is able to identify metabolism-related molecular profile differences between patients with colorectal cancer and healthy people. The analysis...

Fine-Tuned LLMs Boost Error Detection in…

A type of artificial intelligence (AI) called fine-tuned large language models (LLMs) greatly enhances error detection in radiology reports, according to a new study published in Radiology, a journal of...

DeepSeek-R1 Offers Promising Potential t…

A joint research team from The Hong Kong University of Science and Technology and The Hong Kong University of Science and Technology (Guangzhou) has published a perspective article in MedComm...

Deep Learning can Predict Lung Cancer Ri…

A deep learning model was able to predict future lung cancer risk from a single low-dose chest CT scan, according to new research published at the ATS 2025 International Conference...

New Research Finds Specific Learning Str…

If data used to train artificial intelligence models for medical applications, such as hospitals across the Greater Toronto Area, differs from the real-world data, it could lead to patient harm...

'AI Scientist' Suggests Combin…

An 'AI scientist', working in collaboration with human scientists, has found that combinations of cheap and safe drugs - used to treat conditions such as high cholesterol and alcohol dependence...

Patients say "Yes..ish" to the…

As artificial intelligence (AI) continues to be integrated in healthcare, a new multinational study involving Aarhus University sheds light on how dental patients really feel about its growing role in...

Brains vs. Bytes: Study Compares Diagnos…

A University of Maine study compared how well artificial intelligence (AI) models and human clinicians handled complex or sensitive medical cases. The study published in the Journal of Health Organization...

Philips Foundation 2024 Annual Report: E…

Marking its tenth anniversary, Philips Foundation released its 2024 Annual Report, highlighting a year in which the Philips Foundation helped provide access to quality healthcare for 46.5 million people around...