Mia

Kheiron Medical Technologies

Mia supports radiologists in making the recall decision in breast cancer screening. Available to use in various configurations such as an independent second reader, a concurrent reader or in double reader triage. It also offers an AI-enabled breast positioning and image quality assurance tool: Mia IQ.

Information source: Vendor
Last updated: June 9, 2024

General Information

General
Product name	Mia
Company	Kheiron Medical Technologies
Subspeciality	Breast
Modality	Mammography
Disease targeted	Breast cancer
Key-features	Independent second read, recall suggestion, suspicious region marking, image indications
Suggested use	Before: stratifying reading process (non, single, double read) During: perception aid (prompting all abnormalities/results/heatmaps) After: diagnosis verification

Technical Specifications

Data characteristics
Population	Mammography screening population
Input	2D Full-Field Digital Mammography
Input format	DICOM
Output	Recall or no recall suggestion, image indications, auxiliary regions of interest
Output format	DICOM Mammography Structured Report or Presentation State
Technology
Integration	Integration in standard reading environment (PACS)
Deployment	Cloud-based, Hybrid solution
Trigger for analysis	Automatically, right after the image acquisition
Processing time	10 - 60 seconds

Regulatory

Certification
CE	Certified, Class IIa , MDD
FDA	No or not yet
Intended Use Statements
Intended use (according to CE)	Mia is indicated to aid readers in the interpretation of breast imaging examinations for the early detection and diagnosis of breast cancer. The device can serve as a concurrent reader in all reading workflows, including single and/or double reading workflows. Additionally, the device can serve as an independent second or third reader in blinded or unblinded workflows. The device can also be used in the triaging of examinations for workflow prioritisation, and resource management. Output of the device may include classification information such as case-wise suggestion (suspicion of malignancy and/or recommendation that the subject/patient ‘should be recalled’ or ‘should not be recalled’ for further assessment), and localisation information of suspected malignancies such as left and/or right side, projection (e.g. CC and/or MLO views), and region(s) of interest (ROI).

Market

Market presence
On market since	10-2018
Distribution channels	Various
Countries present (clinical, non-research use)	10+
Paying clinical customers (institutes)
Research/test users (institutes)
Pricing
Pricing model
Based on

Evidence

Evidence
Peer reviewed papers on performance	Cost-effectiveness requirements for implementing artificial intelligence technology in the Women’s UK Breast Cancer Screening service (read) Prospective implementation of AI-assisted screen reading to improve early detection of breast cancer (read)
Non-peer reviewed papers on performance	Large-scale evaluation of an AI system as an independent reader for double reading in breast cancer screening (read)
Other relevant papers	Comparison of Mammography AI Algorithms with a Clinical Risk Model for 5-year Breast Cancer Risk Prediction: An Observational Study (read)