AI Model Rivals Dermatologists in Assessing Skin Cancer Aggressiveness

AI Model Challenges Dermatologists in Skin Cancer Assessment

Researchers at the University of Gothenburg have developed a simple AI model that performs on par with experienced dermatologists when assessing the aggressiveness of squamous cell carcinoma, a common form of skin cancer 1

1

. This breakthrough could potentially revolutionize preoperative assessments and surgical planning for skin cancer patients.

The Challenge of Squamous Cell Carcinoma

Squamous cell carcinoma is the second most common form of skin cancer in Sweden, affecting over 10,000 Swedes annually 1

1

. It typically develops in sun-exposed areas and is strongly linked to accumulated UV radiation over time. While diagnosis is often straightforward, the real challenge lies in determining the tumor's aggressiveness before surgery.

AI Model Development and Performance

The research team trained an AI system using 1,829 clinical close-up images of confirmed squamous cell carcinoma cases. The model's ability to distinguish three levels of tumor aggressiveness was then tested on 300 images and compared with assessments from seven experienced dermatologists 2

2

.

Results published in the Journal of the American Academy of Dermatology International showed that the AI model performed almost identically to the team of medical experts. Interestingly, agreement between individual dermatologist assessments was only moderate, highlighting the complexity of the task 1

1

.

Key Findings and Implications

The study identified two clinical features - ulcerated and flat skin surfaces - as clear indicators of more aggressive tumor growth. Tumors with these characteristics were more than twice as likely to fall into higher levels of aggressiveness 2

2

.

This AI model could be particularly valuable in countries like Sweden, where preoperative punch biopsies are not routinely performed for suspected squamous cell carcinoma. The technology offers a non-invasive alternative for assessing tumor aggressiveness, potentially improving surgical planning and patient outcomes 1

1

.

Future Prospects and Challenges

While the use of AI in skin cancer care has generated significant interest, its practical impact on healthcare has been limited so far. Dr. Sam Polesie, the lead researcher, emphasizes the importance of integrating AI where it adds real value to decision-making processes in healthcare 2

2

.

The model developed in this study requires further refinement and testing before clinical implementation. However, it represents a promising step towards more accurate and efficient preoperative assessments for skin cancer patients, potentially reducing the need for invasive procedures and improving overall care quality.