A medical artificial intelligence company, LifeSemantics, has recently announced the successful completion of a clinical study for “skin cancer image detection and diagnostic assistance software.” This marks a significant milestone in the field of digital health, particularly in the fight against skin cancer.
Clinical Study and Outstanding Results
A clinical study was carried out to show the effectiveness of a skin cancer diagnostic tool developed by the Doctor Answer 2.0 project, led by the National IT Industry Promotion Agency (NIPA).
Three hospitals—Kyungpook National University Hospital, Keimyung University Dongsan Hospital, and Yeungnam University Hospital—worked together on this study. They analyzed 199 cases using an algorithm that had collected 6,500 images of skin tumors from patients suspected of having skin cancer.
LifeSemantics used these clinical studies to distinguish between malignant and benign tumors. The results were impressive across all measures, including sensitivity, specificity, and accuracy. The diagnostic tool performed exceptionally well, exceeding the minimum clinical performance standards with a diagnostic accuracy of 80.9%.
The rising importance of skin cancer
According to the statistics from the Health Insurance Review and Assessment Service, the importance of skin cancer is on the rise, with the number of people receiving treatment for skin cancer rapidly increasing from 23,605 in 2018 to 31,661 in 2022. Representative examples include squamous cell carcinoma derived from epidermal keratinocytes, basal cell carcinoma derived from basal cells, and malignant melanoma derived from melanocytes.
In particular, malignant melanoma is a deadly skin cancer that metastasizes rapidly and is difficult to diagnose with the naked eye, accounting for about 65-75% of skin cancer deaths, making early diagnosis very important.
AI Solution for Skin Cancer Detection
LifeSemantics’ skin cancer image detection and diagnostic assistance software is a medical artificial intelligence solution that detects skin cancer and is expected to aid in the early diagnosis and prevention of skin cancer by learning data on basal cell carcinoma, squamous cell carcinoma, and malignant melanoma.
Moreover, as the images taken with smartphones are used without expensive equipment, it reduces the burden for primary medical institutions, which should have a positive impact on the expansion of the introduction of medical AI and the improvement of patient compliance.
Future plans and expectations
LifeSemantics, designated as skin cancer image detection and diagnostic assistance software under the “100 major tasks for food and drug regulatory innovation” by the Ministry for Food and Drug Safety last November, plans to accelerate the licensing and commercialization of medical devices.
Heo Eun-young, the director of the research and development center of LifeSemantics, stated, “This is the first case in Korea to demonstrate the performance of artificial intelligence for skin cancer through clinical studies. With the results of clinical studies conducted abroad, we can increase the accessibility to treatment for skin cancer patients in Korea.”.
Meanwhile, LifeSemantics is conducting clinical studies on “home blood pressure prediction software,” “hypertension complication prediction software,” and “hair density analysis software,” in addition to “skin cancer image detection and diagnostic assistance software.” They expect to complete the clinical studies and licenses within the year and introduce the “CanofyMD” line.
This development is a testament to the potential of AI in revolutionizing healthcare, particularly in the early detection and treatment of skin cancer. It’s a promising step forward in the digital health landscape, bringing us closer to a future where AI plays a crucial role in medical diagnostics and patient care.
