Method for determining severity of skin disease based on percentage of surface area of human body covered by lesion

Abstract

An image processing method for automatically calculating a body surface area (BSA) score using a machine learning technique is provided. The proposed region in each of the plurality of training set images is defined using a Felzenszwalb image segmentation algorithm. The training set image is over-segmented, and each of the proposed regions in each of the plurality of over-segmented training set images is then manually classified as a lesion or a non-lesion. Then, a convolutional neural network (CNN) is trained using the proposed region of the manual classification in each training set image of the plurality of training set images. The trained CNN is then used on a test image to calculate a BSA score.

Description

Background technique [0001] Disease severity assessment for skin diseases such as psoriasis involves calculating the percentage of body surface area covered by lesions and inflammation (ie, BSA score). Hereinafter, lesions and inflammation are collectively referred to as "lesions". BSA is the measured or calculated surface area of ​​the human body. [0002] Psoriasis is an autoimmune skin disease that manifests as red and inflamed areas that differ from healthy normal skin. An important part of disease severity measurement in psoriasis is monitoring what percentage of the body's surface area is covered by inflamed areas called "lesions." For plaque psoriasis, the two main disease measures are BSA and PASI (Psoriasis Area and Severity Index), both of which involve the calculation of a percentage score used to monitor disease progression and treatment effect (ABozek, A .Reich (2017)). Reliability of three psoriasis assessment tools: Psoriasis Area and Severity Index, Body Su...

AI Technical Summary

Problems solved by technology

The main problem with current psoriasis disease scores is that they are imprecise and rough estimates with human bias

Additionally, the process of calculating the percentages to obtain an overall PASI score is tedious and time-consuming

Another clinical need is that currently no BSA measurement exists for guttate psoriasis where the infected body area is large and covered by multiple inflammatory lesions ranging in size from 2mm to 10mm such that Difficult to measure by sight

Images