nodoca throat photos met preset accuracy goals in Japan's review

In Japan's review of nodoca, its AI reading of throat photos detected influenza with 76.0% sensitivity and 88.1% specificity against PCR, meeting goals set in advance.

  • Not rated an organization

    The approved version was tested on stored trial data, and no diagnosis depended on its results.

  • Not rated a person

    Trial patients felt less pain with the earlier device's throat photo than with the swab, but no patient has yet been diagnosed with the approved version.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Intended benefit.
Evidence

foundstudiedreportedclaimed

An independent regulatory document supports the benefit.

Compared with

Sensitivity and specificity goals set before the additional test (the published report withholds the goal values). Measured.

The system’s part
Not isolated. PMDA states that Aillis built the device to detect influenza findings by AI image analysis of the throat. The trial measured the device's accuracy against PCR and did not compare doctors' diagnoses with and without it. The lower pain comes from replacing the swab with a throat photo, which the AI analysis makes usable for diagnosis.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
nodoca (ノドカ)
Made by
Aillis, Inc. (アイリス株式会社)
Used by
Aillis (trial at 11 sites in Japan, with the approved version run on stored trial data)
Where
Japan (11 trial sites)
Stage
Tested in a laboratory
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Identifying Objects, Actions, and Events, Analyzing Data or Information
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Regulatory document, by authors with no tie to the system.

    Pharmaceuticals and Medical Devices Agency (PMDA). 審査報告書 ｎｏｄｏｃａ（ノドカ）, review report on the application by Aillis, Inc., 15 February 2022, published with the deliberation report of the Ministry of Health, Labour and Welfare dated 9 March 2022. Summary of review results and section 4 (総合評価).

    追加試験における感度と特異度の 95%信頼区間の下限値は、それぞれ 70.7%、85.5%であり、事前に設定された性能目標は満たされたため、本品の有効性は示されたと判断した。
    検体採取にともなう痛みの度合いや涙流頻度がイムノクロマト法検査と比べ有意に低かったことから、本品の安全性に関して特段の懸念はないと判断した。

Details

AI system

Workflow

The camera photographed the back of the throat, and the software analyzed the image with 26 items of interview data such as temperature and symptoms. The approved version was run on stored data from 659 patients of a trial at 11 sites in Japan, after Aillis changed the program and the camera.

Follow-up

Conditions for replication

Patients are at least 6 years old and have influenza-like symptoms. A doctor makes the diagnosis with other clinical findings, and the device does not replace the rapid antigen test. Other infections that produce lymphoid follicles in the throat may need other tests.

Entry details

Sector

Healthcare.

Purpose

Aillis built nodoca to help doctors diagnose influenza by analyzing photographs of the throat together with interview data.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). nodoca throat photos met preset accuracy goals in Japan's review (AIBD-0873). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0873.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions