Machine learning-based ABA treatment recommendation and personalization for autism spectrum disorder: an exploratory study
importance 7/10
ai assisted
Autism Spectrum Disorder Research autism therapy stage: pilot
Problem
Shortage of licensed ABA practitioners and subjective, non-data-driven planning limits timely autism treatment.
Why it matters
Rising ASD cases mean many children cannot get timely, quality behavioral treatment.
What the AI did
Patient similarity and collaborative filtering methods recommended and personalized ABA treatment goals for 29 study participants.
Results
Average accuracy 81–84% and NDCG 79–81% versus clinician-prepared ABA treatment recommendations, on 29 participants.
Impact
Feasibility shown on a small exploratory sample; clinical adoption not stated.
Record
| Venue | Brain Informatics |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Manu Kohli, Arpan Kumar Kar, Anjali Bangalore, Prathosh AP |
| Citations | 51 (OpenAlex, at import time) |
| References | 111 |
| Model or method | patient similarity and collaborative filtering |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1186/s40708-022-00164-6 |
| Open access | gold |
| Keywords | autism, ABA, collaborative filtering, patient similarity, treatment recommendation |
Importance rating
7
significant — thousands to millions affected
Abstract states rising ASD cases and a substantial practitioner shortage limiting timely treatment for a large patient population; no numbers given; validated on only 29 participants.