SSA · no agency ID · record R3096

Intelligent Medical Language Analysis Generation (IMAGEN)

Social Security Administration · Chief Information Officer, Disability Information Systems

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: High-impactSource: a) High-impact
Impact justification: Not reported
Start date: 2021-08-20 (day precision)Source: 8/20/2021 0:00
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Government Benefits Processing
Operational functions: Clinical and medical Derived by keyword rules; see methodology

Problem the AI is intended to solve

This AI use case helps employees visualize, search, and more easily identify relevant clinical content in medical records. The use case improves efficiency and accuracy of disability determinations/decisions.

Expected benefits

The use case improves efficiency and accuracy

System outputs

Document annotations on medical records to aid in navigation such as identifying patient encounters, tests, dates, and other relevant information. Possible impairment and SSA Medical Listing Codes are also identified.

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): IMAGEN
Custom-developed code: Yes
Public source code: Not reported

Sourcing

How it was built: Developed in-houseSource: b) Developed in-house
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: Yes
Privacy Impact Assessment: Not reported
Authorization to Operate: Yes
Demographic features: c) Age
Training and evaluation data: Historically adjudicated disability claims and outcomes. Manually annotated documents
Federal Data Catalog: Not reported

Governance

8 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingReported in placeSource: a) Yes

AI impact assessmentReported in placeSource: a) Yes

Independent reviewReported in placeSource: c) Yes – by the CAIO

Ongoing monitoringReported in placeSource: a) Yes, sufficient monitoring protocols have been established

Operator trainingReported in placeSource: a) Yes, sufficient and periodic training has been established

Fail-safeReported in placeSource: a) Yes

Appeal processNot applicableSource: b) Not applicable

User and public consultationOther or ambiguous answerSource: d) Other

Potential impacts and how they were identified

The risks include overreliance on system-generated findings and possible misinterpretation of results due to system limitations. These impacts were identified through collaborative team discussions, manual case reviews, user feedback, and ongoing monitoring of system performance and usage. The assessment process included evaluating known limitations, potential harms, and failure modes. However, notably, the system never suggests an adverse outcome for a disability determination.