HHS Raises TOPx Prize Pool as AI Prototypes Target Lyme Disease and Invisible Illnesses
The U.S. Department of Health and Human Services has expanded the TOPx HHS Tech Sprint for AI and Invisible Illness prize pool from $2 million to $3.8 million and awarded $3.7 million to 40 Phase 2 winners developing working prototypes for complex chronic conditions. The competition sits at the intersection of diagnostics, care coordination, and the growing use of artificial intelligence to make chronic disease burden more visible in clinical and policy settings.
For Lyme disease, that visibility problem is not abstract. Patients can move through prolonged diagnostic delays, fragmented specialty care, and symptoms that are difficult to quantify in routine encounters. AI tools built on federal open data are being asked to bridge that gap by organizing clinical signals, improving communication across care teams, and translating lived experience into a format that clinicians and systems can act on more quickly.
The prize structure itself reflects the breadth of the challenge. Two grand prizes of $1 million each went to LasaClinical: Reducing the Diagnostic Delay in Women’s Health and Health Worth. Five teams received first-place awards of $200,000 each, six teams earned second-place awards of $50,000 each, and 27 teams received $15,000 impact awards. A further $50,000 in People’s Choice Awards will be distributed later after public voting and final approval.
A competition that drew 171 submissions and advanced 46 teams into Phase 2 is not just a technology exercise; it is a signal of how much unmet need remains in the diagnosis and management of complex chronic illness. In practical terms, these tools are trying to solve one of medicine’s most persistent failures: converting diffuse symptoms, uncertain histories, and incomplete documentation into a clearer clinical pathway.
Diagnostic Delay and Care Coordination at the Center of the Effort
The announced winners were asked to build tools across three challenge areas: Lyme Innovation, Invisible Illness, and Cost of Illness. Each area maps to a different failure point in modern care. Lyme Innovation addresses the difficulty of identifying disease early and accurately. Invisible Illness focuses on conditions whose symptoms are real but often poorly captured by standard metrics. Cost of Illness aims to quantify the economic and social consequences that are frequently undercounted when chronic disease is reduced to a narrow set of billing codes or visit notes.
The biological and clinical logic behind these efforts is straightforward. When diagnosis lags, treatment decisions are delayed, symptom burden can intensify, and patients may accumulate avoidable visits, testing, and downstream disability. AI systems that can organize complex histories, surface patterns from open datasets, and support navigation across providers may help shorten some of that delay. They will not replace bedside judgment, but they may reduce the administrative and informational friction that often stands between a patient and a coherent evaluation.
In the words released with the announcement, the overarching goal is to shorten the path to diagnosis, improve care, and help patients reclaim health. Those are broad aims, but the operational target is narrower: turn disparate data into usable tools for clinicians and patients who are currently forced to assemble their own care plans across disconnected systems.
Federal Data, Nonfederal Review, and the Push for Working Prototypes
The challenge was led by the HHS LymeX Innovation Accelerator in collaboration with the National Institutes of Health Office of Research on Women’s Health and the Federal Chief Data Officers Council. Submissions were reviewed through nonfederal expert evaluation and federal review using published criteria. That structure matters because the competition was not simply rewarding concept papers. It was selecting working prototypes that could be taken from idea stage toward practical deployment.
The emphasis on federal open data is especially relevant. Open datasets can help developers train and test AI tools without depending entirely on proprietary sources that may be narrow, incomplete, or poorly representative of the patients who most need help. In a domain such as Lyme disease, where symptom patterns may be variable and clinical histories are often fragmented, the value of broader public data is clear: it can support tools designed to spot patterns earlier, organize longitudinal information, and make case complexity more legible to busy clinicians.
Public reporting also framed the award announcement as part of a wider Lyme disease initiative tied to a presidential executive order directing HHS to build on the Lyme Moonshot through prize competitions or research and development. Policy support of that kind can accelerate innovation, but it also raises the bar for what comes next. Prototype-stage success does not equal clinical effectiveness, and no prize competition by itself establishes that a tool improves diagnostic accuracy, patient outcomes, or cost of care.
What the Awards Can and Cannot Deliver Yet
The most immediate impact of the TOPx awards may be organizational rather than therapeutic. Tools that help patients explain symptoms, help doctors interpret prolonged illness, or help systems see the cumulative cost of delayed diagnosis could change the flow of care long before they change a biologic endpoint. For patients with Lyme disease and related invisible illnesses, a more legible record of symptoms and care needs could reduce the burden of repeatedly telling the same story without being believed or adequately evaluated.
Still, the clinical value of these prototypes will depend on implementation. An algorithm that organizes data well in a competition setting may struggle in a real clinic if it is slow, hard to integrate, or unable to accommodate the variability of individual patients. Equally important, any AI tool used in this space must avoid amplifying bias by treating incomplete data as certainty. Complex chronic illness already suffers from under-recognition; poorly designed digital tools could compound that problem rather than solve it.
The competition’s size and award distribution show that HHS is casting a wide net across diagnostic support, patient navigation, and cost visibility. What remains unknown is which prototypes will prove durable in real-world use, how they will be validated, and whether they can meaningfully change the time it takes for patients with Lyme disease or other invisible illnesses to receive credible answers. Those questions will determine whether the prize program becomes a policy exercise or a genuine clinical toolset.