Computer aided diagnosis of melanoma
Project Leader: Mohammad Aldeen
Collaborators: Rahil Garnavi (IBM R&D Australia), Goerge Varigos (Royal Melbourne Hospital)
Primary Contact: Mohammad Aldeen (firstname.lastname@example.org)
Keywords: artificial intelligence; bioinformatics; biomedical engineering; cancer; machine learning
Disciplines: Electrical & Electronic Engineering
Domains: Convergence of engineering and IT with the life sciences
Malignant melanoma is the deadliest form of skin cancer. In Australia, melanoma is the most common cancer in people aged 15–44 years. It represents10% of all cancers and its per-capita incidence is four times higher than in Canada, the UK and the US, with more than 10,000 cases diagnosed and around 1250 deaths annually. The worldwide steady increase in incidence of melanoma in recent years, its high mortality rate and the massive respective medical cost has made its early diagnosis a continuing priority for public health. Early diagnosis of melanoma is particularly important for two reasons. First, the prognosis of melanoma patients highly depends on tumour thickness. If melanoma is detected at early stages, it is highly curable, with a 10-year survival rate between 90 and 97%. Due to the enhancements in skin imaging technology and image processing techniques in the recent decades, there has been a significant increase in interest in the computer-aided diagnosis of melanoma. The aim is to remove subjectivity and uncertainty from the diagnostic process, provide a reliable second opinion to dermatologists and overcome the low reproducibility found in clinical diagnosis. Through digital representation of dermoscopy images and in-depth mathematical and statistical analysis of colour and intensity of the lesion, it is expected that the computer algorithm (software) recognises features that are not detectable by human eyes, and therefore improves the diagnostic accuracy of even well trained and experienced clinicians.
Vitamin D Digital Meter: This project, which is a joint venture with the Royal Melbourne Hospital, aims at designing and constructing a device that measures the amount of vitamin D a human body absorbs when exposed to the Sun. The device reads and records the total exposure time, time of day, sun intensity level, and UV profile. All this information is processed by a micro-processor to calculate the amount of vitamin D absorbed based on clinical data provided by the Department of Dermatology at the Royal Melbourne Hospital.
Psoriasis Index Valuator: Psoriasis is a disease that affects the human skin causing red scaly spots. The most widely used measure of severity used by dermatologists is a score called PASI (Psoriasis Area Severity Index), which is a subjective measure of colour, texture and depth of affected areas. This index is assessed through visual examination and thus relies heavily on the expertise of the physician. To remove subjectivity, signal and image processing techniques are being investigated to provide consistent and highly accurate alternative to visual examination by dermatologists.
3D Mapping of Human Body: This project involves the mapping of human body onto a 3-dimenstional template so that any lesion on the skin is accurately mapped and its statistics recorded for future reference by dermatologists. Some of these lesions may not be detectable by human eyes, and therefore this technology improves the diagnostic accuracy of even well trained and experienced clinicians.
Further information: http://www.findanexpert.unimelb.edu.au/researcher/person15099.html