Monitoring the ‘human machine’ with AI

Artificial intelligence (AI) is a rapidly evolving field, and more and more useful AI-based tools and technologies are being developed for the medical world and dermatology clinical practice. Dr Alexander Zink (University of Munich, Germany) provided an update with respect to the latest developments in the field [1].

“We are increasingly approaching the human body as a machine that is continuously monitored,” said Dr Zink. “Face masks that can measure breathing patterns, smart textiles that can measure skin temperature and metabolites, and smartwatches that can track activity, sleep, heart rate, and blood oxygen levels are just some of the devices that have been developed or are in the pipeline for monitoring the ‘human machine’ [2].” Integrated wearable microneedles to monitor biomarkers in the interstitial fluid, wireless patches to track C-reactive protein in sweat, and a wearable nano-biosensor to monitor female hormones are some other novel monitoring options. “But we don’t want one wearable for diabetes, and another for cardiovascular issues; ideally, we would have one solution for all of it,” according to Dr Zink. One innovation aiming to monitor many parameters simultaneously is the development of nucleic acid-based wearable and implantable electrochemical sensors. These could be adjusted to the individual’s needs, measuring biomarkers of interest to track certain health conditions [3].

Dr Zink argued that this monitoring will not stop at the body but will extend into an individual’s environment. “There are mattresses that can track nocturnal scratching in atopic dermatitis and refrigerators that can track dietary habits in patients with hidradenitis suppurativa [4],” he said. “Also, an innovative smart mirror can analyse the skin, providing personalised skin care recommendations.” In addition, this tool can non-invasively track blood sugar by observing the colour of a person’s lips [5]. Furthermore, smart toilets capable of analysing urine and stool, as well as conducting other measurements, are currently in development [6].

In dermatology, these monitoring tools and innovations are particularly helpful when used alongside a dermatologist's assessment. It has been demonstrated that the diagnostic performance of a dermatologist working in collaboration with a convolutional neural network is superior to the diagnostic performance of either one alone [7].

Other promising applications of AI in dermatology include communication tools, such as answering patients’ questions about their condition. “A panel of dermatologists gave very high ratings to ChatGPT’s answers to common patient queries, such as ‘Can you tell me more about the cause of my eczema?’ or ‘How can I treat eczema flare-ups?’” [8]. Moreover, ChatGPT was considered to be more empathetic than physicians [9].

“Today we still compare patients based on demographic features and general characteristics,” said Dr Zink. “However, this is not a very precise method for patient stratification. For example, Prince Charles and the late Ozzy Osbourne are both males from 1948, raised in the UK, who have been married twice, and lived in castles. Nevertheless, they are completely different individuals with divergent lifestyles, illustrating the limitations of such comparison.”

AI is playing an increasingly important role in medicine, offering communication tools, monitoring instruments, enhancing diagnostic accuracy and other advantages. Importantly, the best outcomes may come from a collaboration between clinicians and AI-based tools, highlighting that AI is not necessarily a replacement for human intervention, but rather an augmentation of the clinical toolkit. At least for now…

1. Zink A. Artificial Intelligence. SPIN Dermatology Congress, Paris, France, 2-4 July 2025
2. Ates HC, et al. Nature Electronics. 2021;4:13-14
3. Ye C, et al. Chem Society Reviews. 2024;53:7960-7982
4. Gambhir SS, et al. Sci Transl Med. 2018;10(430):eaao3612
5. https://gagadget.com/en/559131-samsung-shows-off-the-micro-led-beauty-smart-mirror-it-can-analyse-skin-and-give-care-recommendations/ visited 5 August 2025
6. Jessie GE, et al. Sci Transl Med. 2023;15(681):eabk3489
7. Winkler JK, et al. JAMA Dermatology. 2023;159(6):621-627
8. Sulejmani P, et al. JEADV. 2024; 38(6):e531-e535
9. Ayers JW, et al. JAMA Intern Med. 2023;183(6):589-596

Copyright ©2025 Medicom Medical Publishers



Posted on 21 August 202521 August 2025