Abstract Introduction Rates of notifiable Sexually Transmitted Infections (STIs) have been escalating in Australia over the last ten years. In 2013 there were 102 000 notifiable STI cases, including Chlamydial, Gonococcal and Syphilis; this has steadily climbed over the previous ten years to 159 000 cases in 2023. In particular there have been higher rates of STIs in the Northern Territory and Queensland states; with 2302 cases per 100 000 and 650 cases per 100 000 respectively. In recent years there has been a lot of discussion about the role and use of Artificial Intelligence (AI) in healthcare. With developing technology and improved image analysis, the role of AI software in healthcare and aiding in the diagnosis of dermatological conditions has been a hot topic. Objective Access to and engagement with Australian primary healthcare services for assessment of a possible STI is fraught with difficulty for a myriad of reasons. Access to General Practitioners (GP) and primary health care within Australia has been very difficult in recent years due to a shortage of GPs throughout the country. This has had a particularly significant effect in states such as the Northern Territory and Queensland which have large geographical footprints and many rural communities that have traditionally struggled to retain local GPs, and often have to drive for greater than 3 hours to reach an alternate medical facility (primary care or emergency department). STIs are also typically stigmatised and shamed which further discourages individuals to seek medical help. Methods On review of the literature; AI image analysis has been shown to have been of benefit for diagnosis and surveillance of dermatological conditions such as; skin cancer, psoriasis, atopic dermatitis etc. It has been shown to be of use as an adjunct to the clinician, increasing the sensitivity and accuracy of screening for these dermatological conditions. Results As discussed above, STIs are on the rise within the Australian population over recent years. We propose that AI image analysis could assist with the assessment and diagnosis of STI in the Australian community in the following ways. Risk stratification using non-AI software to triage STI risk based on known risk factors; such as unprotected sexual activity, number of sexual partners, previous STIs etc. AI image analysis to aid with diagnosis of STIs; as per the dermatological examples this would assess a picture taken of genital lesions and compare to a data-base to determine likelihood a STI and triage those that require urgent clinician review (Noting that this would need to be a secure database given the intimate nature of the images). This could also be linked into automated formal STI pathology request generation if the AI image analysis is of high concern for an STI (and the AI program has been shown to have good sensitivity and accuracy) to aid in early diagnosis and treatment. Conclusions In particular the use of an AI image analysis service would be of greater benefit in large rural communities such as the Northern Territory and Queensland, where access to medical care can be limited. Disclosure No.
Read full abstract