Digital Analysis and Lesion Segmentation Software for Tele-Dermatology
T2018-031 An automated image analysis system for the recognition and quantification of skin disease.
Acne and rosacea are skin diseases that affect around 85% of individuals. There is no gold standard for evaluation of these skin diseases, and their treatment efficacy is generally determined according to a poorly validated counting process by the physician and patient. This process is prone to poor reliability between different observers and can be highly time consuming, making the assessment challenging and inefficient.
Acne can be classified into several morphologies including closed comedones (whiteheads), open comedones (blackheads), papules, pustules, cysts (nodules), and scars. For a validated assessment, the different morphologies need to be counted independently. This is not cost-effective in most settings; thus, a general gestalt needs to be used clinically. Given that these diseases are highly prevalent, the current methods are cost-prohibitive and suffer from poor reliability. Accordingly, there is a practical need to automate and digitalize the process of evaluating acne and rosacea lesions.
Researchers at The Ohio State University have created novel image analysis software for the recognition and stratification of various skin lesion morphologies and redness in skin disease. This software is ideal for monitoring patient progression during dermatological clinical trials where this technology could decrease cost, investigator time, and inter-rater variability between investigators. The novel software uses a series of time-stamped digital photographs captured with plain white light and readily accessible smart phone cameras to automate lesion identification and quantification. The software automatically adjusts for variance in lighting and generates a map of the lesions. Using proprietary algorithms and methodologies, the skin lesions are scored and classified into six different lesion morphologies for an accurate patient assessment. A summative global score is then calculated for each patient's time-stamped entry to quantify efficacy of clinical treatments.
- Generates quantitative metrics for clinical trials of rosacea or acne treatments
- Reduces clinical trial costs by reducing investigator time and enabling remotely captured data points without a standard office-visit for counting
- Results correlate strongly with expert consensus and repeated analysis
- Reduces paper documentation and physician notes of the evaluation
- Harmonizes data across participating clinical trial physicians and/or institutions
- Can improve medication patient adherence by precise quantification of change for patients remotely
- Dermatology clinical trial patient scoring and data management
- Currently applicable to rosacea, acne, and psoriasis
- Patient-facing smart phone application development, e.g.. Grade Your Acne, Count Your Acne Over Time
- Pharmaceutical company development for patient-facing smart phone applications to encourage adherence with their drug, e.g. See your improvements over time graphed using XX pharmaceutical
- International Patent Application No. PCT/US2018/054054