How a Saudi university is using AI to transform skin disease diagnosis and treatment
RIYADH: To make dermatological care more efficient and effective, experts at King Abdullah University of Science and Technology in Saudi Arabia have developed a groundbreaking diagnostic system, SkinGPT-4, that harnesses the power of artificial intelligence.
Xin Gao, professor of computer science, co-chair of the Center of Excellence in Smart Health and head of the Bioinformatics Platform at KAUST, who led the research, said the goal of SkinGPT-4 is to detect, diagnose and determine appropriate treatments for skin diseases.
Gao says the technology, developed in collaboration with Juexiao Zhou, a PhD student at KAUST and first author of SkinGPT-4, has the potential to provide a life-saving service to patients, especially those in rural areas where there is often a shortage of trained dermatologists.
“These specific challenges in dermatology led to the creation of SkinGPT-4,” Gao told Arab News. “The variability of skin presentation and the need for specialized knowledge to correctly identify and treat these conditions highlighted the need for an advanced AI-based solution.”
The team identified the need for such a solution after recognizing the limitations of traditional diagnostic methods and the potential of artificial intelligence, especially large language models (LLMs) such as the eponymous ChatGPT, in improving the accuracy and efficiency of dermatological diagnostics.
“With SkinGPT-4, users can upload their own skin photos for diagnosis, and SkinGPT-4 can autonomously determine the characteristics and categories of skin diseases, perform analysis, provide treatment recommendations, and enable interactive diagnosis,” Gao said.
Gao and the KAUST team behind the AI-based dermatologist SkinGPT-4. (Supplied)
SkinGPT-4 diagnoses conditions with distinctive visual features such as acne, rosacea, melanoma, psoriasis, basal cell carcinoma, eczema, and many others.
Gao said the development of SkinGPT-4 began with data collection and preprocessing, followed by model training and validation. “The team collected a large dataset of dermatological images and medical records to train the AI model,” he said.
“One major challenge was integrating different types of data, including images and text, which required collaboration between computer scientists and dermatologists. A multidisciplinary team worked together to ensure that the AI could effectively interpret and analyze images of skin diseases.”
DO YOUKNOW?
• SkinGPT-4 diagnoses conditions with specific visual features, such as melanoma, psoriasis and eczema.
• Uses a combination of computer vision algorithms, large language models, and natural language processing.
• This technology has the potential to help doctors and patients in rural areas where there is often a shortage of qualified dermatologists.
SkinGPT-4 uses a combination of computer vision, LLM, and natural language processing (NLP) algorithms that enable programs to understand human language.
“The model processes dermatological images using a vision transformer (ViT) to identify patterns and features that indicate various skin conditions,” Gao said.
“ViT is compatible with the LLM named Llama-2-13b-chat on our dataset with an adapted two-step training strategy. This enables the LLM Llama-2-13b-chat to understand skin disease images and enable conversational diagnosis with the patient in natural language.”
This section contains important reference points placed in (opinion field)
The SkinGPT-4 test may prove particularly useful in diagnosing rare skin conditions that family doctors may have difficulty recognising.
“A patient coming in with an unusual rash can get a quick and accurate diagnosis with SkinGPT-4, which has been trained on a wide range of dermatological images, including rare conditions,” Gao said.
“In addition, when treating chronic skin conditions such as psoriasis, SkinGPT-4 can monitor progress and response to treatment, providing ongoing support and adjusting treatment plans as needed.”
Scientists hope SkinGPT-4 will prove to be a breakthrough in remote or underserved areas where there is a shortage of dermatologists.
“For example, in a rural community where the nearest dermatologist is hundreds of miles away, a patient comes in with a suspicious lesion that could be a rare form of skin cancer,” Gao said.
Properly trained, AI can provide doctors with enormous assistance. (Photo: Shutterstock)
“With SkinGPT-4, a local healthcare provider can take a high-resolution image of a lesion and enter the patient’s medical history into the system. SkinGPT-4 analyzes the image and patient information to quickly provide a preliminary diagnosis and recommendations for further action.”
Gao added that as SkinGPT-4 develops, the system will learn from its mistakes through continuous learning and feedback mechanisms.
“By analyzing misdiagnoses and making corrections, the system can refine its algorithms and increase accuracy over time,” he said. “This iterative learning process ensures that SkinGPT-4 evolves and adapts to new data and emerging trends in dermatology.”
Gao emphasizes, however, that SkinGPT-4 is not intended to completely replace dermatologists. Rather, the program is intended to serve as an evolving and optimizing tool, acting as an assistant to facilitate communication between patients and doctors.
“Our goal with SkinGPT-4 is to provide patients with more information about skin diseases and also to offer physicians valuable assistance in the diagnostic process.”
