AiCure
AiCure's H.Code platform verifies how clinical trial participants take their medication between visits. A participant records each dose on a smartphone, computer vision confirms the dosing event, the app gives dosing guidance, and the study team steps in when a dose is missed or an event needs review. Site dashboards show adherence participant by participant, sponsors follow patterns across sites, and predictive models flag participants at risk of missed doses or early withdrawal. The same app captures patient reported outcomes and records video and audio for digital biomarkers such as hand movement, facial expression and vocal acoustics. Clinical Site Services adds a central monitoring team that follows up with participants alongside site coordinators.
The platform has captured more than 1.5 million participant doses across more than 2,200 trial sites in 46 countries and 45 languages, for seven of the ten largest pharmaceutical companies, with 92% participant retention. Kyle Holen of AbbVie reports dropout cut by more than half, and Cerevel Therapeutics and Sage Therapeutics are among its sponsors. Across 28 trials with more than 4,000 participants in 43 countries, adherent participants in programs using H.Code and Clinical Site Services terminated early at a rate of 7.8%.
H.Code is sold to pharmaceutical and biotechnology sponsors, contract research organizations and sites for traditional, decentralized and hybrid studies, configured for each trial after a demo. Sponsors set which features and data a study uses, who has access and how long data is kept. Sabina Chadha is chief executive of the Houston company.
Capability Axes
An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read
Computer vision assesses each dosing event a participant records on a phone, and predictive models flag patterns tied to missed medication and early withdrawal. Further models read recordings for hand movement, facial expression and vocal acoustics to build study measures. Around them sit electronic outcomes questionnaires, messaging, reminders and site dashboards, with Clinical Site Services adding human follow up through coordinators. Remote dose verification is what sponsors buy, and without the vision and risk models H.Code would be a trial app with reminders and questionnaires.
A participant follows the dosing directions on the phone while the app observes the event, and site staff see adherence and pending events on the dashboard. Access to stored full facial images is limited to trained reviewers. The Sage Therapeutics protocol shows the escalation path at work: the site follows up when a dose is missed or data are pending, can reclassify a dose taken outside the app as site reported, and raises ongoing adherence problems with its medical monitor. Escalation runs through each study's protocol rather than a product rule, and no confidence threshold hands an uncertain recording to a reviewer.
H.Code combines smartphone computer vision for dosing events with models that estimate future adherence and early termination risk, and its digital biomarker tools analyze movement, facial expression and vocal acoustics. Sponsors can receive annotation and model development tools and use de identified recordings for benchmarking, which lets a study team tune its own measures. The computer vision model carries no named version, the clinical risk predictions no description of their training data, and no release history ties a change in behavior to an update.
Smartphone recordings feed H.Code's computer vision and trial analytics, and full facial images stay on AiCure's secured servers with access limited to trained reviewers. Under the European trial notice AiCure processes sponsor data on the sponsor's instructions and protects transfers to the United States and onward to its agents.
That settles who controls the study data, not who handles an inference request: no outside company is named for speech, video verification or risk modeling, and which processing stages stay inside AiCure's own infrastructure is not stated.
AiCure analyzed early termination and adherence across 28 phase I to IV trials with more than 4,000 participants in 43 countries, and adherent participants in programs that combined H.Code and Clinical Site Services terminated early at a rate of 7.8%. The analysis links adherence to withdrawal without a comparison arm that isolates the software.
Kyle Holen of AbbVie reports dropout cut by more than half, with no trial count or design, and the 92% participant retention figure comes without a denominator. No independent prospective study of the current H.Code release accompanies any of them.
Sponsors set collection, access and retention for each covered trial. Under the European trial notice, recordings and identifiable data are deleted at the end of the trial unless regulations require longer retention. Full facial images are encrypted, kept on AiCure servers and open only to a small group of trained reviewers, and H.Code also offers de identified audio and video to sponsors for benchmarking. Collection and storage are covered; model use is not. There is no categorical bar on training with sponsor material, no deletion schedule for model inputs and no defined process for a safety incident such as a wrong verification during a dosing event.
Protected health information moves through the clinical trial app, stored encrypted under AiCure's HIPAA, GDPR and electronic records compliance commitments. The trial subject privacy notice treats the sponsor as controller for covered trials and AiCure as processor where GDPR applies, and sponsors choose what data the system collects, who may see it and how long it is kept, subject to consent and the study protocol, so the privacy terms reach the trial service itself. No business associate agreement is public, and business associate status is not stated for any subscription or deployment.
ISO 27001 and SOC 2 are the frameworks named for the information security program, with encryption at collection, in transit and at rest, and stored full facial images held on secured servers that only a small group of trained reviewers can open. Sponsors decide wider access to a study's data. SOC 2 material goes only to eligible recipients on request, so the report's date, type and scope, the ISO certificate and the services each audit covers are not open to a buyer before asking.
H.Code collects adherence and participant measures for clinical research under a sponsor's protocol. The medication and schedule come from the protocol and investigator, and the app reminds participants and captures how they follow those instructions. Records handling is positioned under 21 CFR Part 11 for electronic records, and the trial privacy notice covers disclosures to FDA in regulatory investigations. H.Code carries no clearance and claims no indication for diagnosing or treating a condition, which fits a research data collection tool sold for use inside a protocol.
H.Code gives sponsor data teams templates, annotation and tools to develop and monitor the models used in a trial, and AiCure's approach to study design calls for validating each model for its intended role, controlling data collection and explaining how a measure drives operational decisions. Trained reviewers control access to full facial images.
The platform runs in 46 countries and 45 languages, yet no result shows whether dose verification performs alike across skin tones, lighting or phone cameras, and there is no subgroup analysis, independent model audit or AI governance certification for the current product.
A trial site can revisit a participant's adherence history and document why a dosing event is missing or misclassified. In the Sage Therapeutics protocol, staff reclassify an unlogged dose as site reported after following up with the participant. The application terms disclaim warranties on the accuracy and completeness of results and put verification on the user, and the trial privacy route handles requests about personal data. The terms carry no performance guarantee, no duty to correct a wrong AI classification and no remedy for the participant.
The participant app sends dosing records and outcomes to site and sponsor dashboards, and H.Code can bring further data sources, companion apps and records into a study workflow with implementation help. In the Sage Therapeutics protocol, staff work from the AiCure dashboard and record reasons for missed doses in the study's source documents by hand. No production interface links H.Code to a named electronic health record, and no interoperability standard or bidirectional exchange is stated for one.
Participants use the app on a personal or study supplied phone, and study teams work from dashboards. Full facial images are stored on AiCure's encrypted servers with access restricted to reviewers. The European privacy notice provides for transfer of trial data to the United States under contractual clauses or the applicable data privacy framework, AiCure can help sponsors with local country filings, and sponsors control collection and retention. No region map covers every deployment, no description separates one sponsor's data from another's, and the analysis service has no customer hosted version.
Pricing is set study by study for pharmaceutical sponsors, contract research organizations and sites. A study can combine smartphone dosing support, outcomes questionnaires, predictive engagement and digital biomarkers, with Clinical Site Services available for work with site coordinators, and the implementation depends on the trial's requirements and complexity. The purchase starts with a demo, followed by a choice of features and support for a particular study. No subscription price, per participant or per site fee or priced bundle lets a sponsor size the cost before a sales conversation.
Trials of every design use it, site based, decentralized, virtual and hybrid, with sponsors collecting measures of patient behavior at home between visits. Its therapeutic areas include neurodegenerative conditions such as Parkinson's and Alzheimer's, psychiatric conditions such as schizophrenia and depression, and infectious disease indications including hepatitis C. The adherence analysis spans phase I to IV studies, and the Sage Therapeutics phase III protocol puts the app and dashboard to work on missed medication in a study of cognitive impairment in Huntington's disease. Validation covers adherence and withdrawal overall, not each therapeutic area or outcome instrument.
Alternatives to AiCure
Other products that do the same kind of work as AiCure, and what each one does.
CluePoints provides clinical trial data quality and oversight software for pharmaceutical sponsors and contract research organizations.
Palo Alto company whose clinical trial platform runs electronic outcome assessment, electronic consent, televisit and participant apps for decentralized, hybrid and site based studies, now with an agent platform for sponsors.
ConcertAI's Accelerated Clinical Trials platform helps pharmaceutical sponsors and contract research organizations design, plan and run oncology studies.
Inato connects clinical trial sponsors with research sites and helps site staff find patients for studies, across more than 6,000 sites in more than 50 countries and more than 25 sponsors, Sanofi, Amgen, Pfizer, Lilly and AstraZeneca among them.
Biorce sells Aika 2.0, an AI operating system for clinical development, to pharmaceutical and biotechnology sponsors and contract research organizations.
Unlearn, based in San Francisco, builds Digital Twin Generators: disease specific machine learning models trained on historical patient level data that forecast how each clinical trial participant would progress under control or standard of care, from their baseline data.
Pricing
Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.
| Entry Price | Pricing Basis | BAA Tier | Implementation | Source |
|---|---|---|---|---|
|
By quote
|
AiCure sells H.Code to research sponsors and study teams through an enterprise demo. Trial complexity, selected modules and the level of site support shape the engagement. There is no fixed rate per participant, site or study. | AiCure processes covered trial data on sponsor instructions. A business associate agreement and a commercial tier for it are unavailable. | AiCure supports study onboarding and workflow integration. Implementation time varies with the trial, and fees depend on the sponsor agreement. | Vendor Published |
H.Code combines dosing verification, outcomes questionnaires, analytics and study dashboards. Sponsors can include Clinical Site Services, study specific data integration and support. Standard minimums and software subscription rates are unavailable.