Technion scientist Hossam Haick is developing a low-cost skin patch that monitors health signals and is designed to detect disease. Commercial use could begin within two or three years, he said. A tuberculosis-detection version cost $1.50 to produce. In a trial of about 2,000 people in India, South Africa and Latvia, it detected active tuberculosis with approximately 90% accuracy compared with a control group.
Six sensors in the patch measure substances released through the skin and send readings to a phone app. Haick said the device still needs testing in real-world conditions to show it can reliably identify disease. He expects the first version to monitor a small number of diseases; later versions could detect more conditions and support personalized treatment.
A sensor technology with security implications
Haick, a professor of chemical engineering and nanotechnology at the Technion, said he faced threats and harassment because he was “an Arab working on sensors, a technology that can be used in missiles.” During the Gaza war, while serving as dean, he moved some exams to the Technion’s Tel Aviv campus because missile fire hit the north. He said threats to his life followed: “They told me, 'We'll slaughter you if you don't change your decision.' Those exact words.” The base account doesn't identify who made the threats.
Haick said people repeatedly humiliated him at Ben-Gurion Airport. He eventually received a “good Arab” certificate, a general fast-track passport control certificate issued by the Israel Airports Authority. In some years, he took 53 flights. He grew up in Nazareth in a Christian family and returned to Israel before the Second Lebanon War, when he said he hid under his desk more than once.
The Bill & Melinda Gates Foundation funded testing of the tuberculosis patch. Haick said the team first made a separate $1.50 version after he had promised Bill Gates a $1 manufacturing cost. The patch can work for months, he said. His team also developed a polymer version that repairs cuts or scratches in about 30 seconds, without heat, ultraviolet radiation or another external stimulus.
Disease detection and validation
Haick’s earlier work focused on identifying volatile organic compounds in exhaled breath. His team built an electronic nose that combines sensors and artificial intelligence to analyze breath for disease markers. The latest device is about the size of a hand and, Haick said, about 88% accurate. It can identify 17 diseases, but the U.S. Food and Drug Administration has granted it Breakthrough Device designation for only two: lung cancer and advanced liver fibrosis unrelated to alcohol consumption.
Hospitals around the world are using the device experimentally to diagnose those conditions. Studies have found it can detect lung cancer early, sometimes before clear symptoms appear. Prof. Jonathan Sleeman of Heidelberg University said clinical validation needs large patient groups, including to establish whether the signals are specific enough to distinguish diseases. “The arguments he presents are compelling, but that has to be shown in the real world, and that's where we're going,” Sleeman said.
Haick is also leading large trials involving tens of thousands of people to identify chemical signatures linked to diseases including malaria, COVID-19, Parkinson’s and several cancers. One lung cancer trial involves 5,600 participants and 20 partners, backed by a 14.8 million-euro grant. He said patches for lung cancer, tuberculosis and hormonal changes in women could reach commercial use within a few years. Clinical validation, regulation, industrial production and market introduction will take longer.
Regional AI research
In Lebanon, the UAE-backed “One Million Lebanese AI Experts” initiative aims to familiarize up to one million Lebanese with generative AI. Minister of the Displaced and Minister of State for Technology Affairs Kamal Sehahdeh said the project, which is still in development, plans three-hour training sessions in four modules. Digital rights advocacy group SMEX has raised concerns about the initiative.
In Abu Dhabi, Insilico Medicine researchers developed LongevityBench, a benchmark for testing AI reasoning across biological data related to aging. Published in Cell in September 2026, it covers 17 tasks. The researchers tested systems from OpenAI, Google, Anthropic, xAI, DeepSeek and Moonshot AI, along with five specialized models. Alex Zhavoronkov, founder and CEO of Insilico Medicine, said the specialized models matched or exceeded most frontier systems. He stressed that AI predictions and aging clocks generate hypotheses; they don't prove efficacy or safety in humans.