Title |
Applying Machine Learning to Facilitate Autism Diagnostics: Pitfalls and Promises
|
---|---|
Published in |
Journal of Autism and Developmental Disorders, October 2014
|
DOI | 10.1007/s10803-014-2268-6 |
Pubmed ID | |
Authors |
Daniel Bone, Matthew S. Goodwin, Matthew P. Black, Chi-Chun Lee, Kartik Audhkhasi, Shrikanth Narayanan |
Abstract |
Machine learning has immense potential to enhance diagnostic and intervention research in the behavioral sciences, and may be especially useful in investigations involving the highly prevalent and heterogeneous syndrome of autism spectrum disorder. However, use of machine learning in the absence of clinical domain expertise can be tenuous and lead to misinformed conclusions. To illustrate this concern, the current paper critically evaluates and attempts to reproduce results from two studies (Wall et al. in Transl Psychiatry 2(4):e100, 2012a; PloS One 7(8), 2012b) that claim to drastically reduce time to diagnose autism using machine learning. Our failure to generate comparable findings to those reported by Wall and colleagues using larger and more balanced data underscores several conceptual and methodological problems associated with these studies. We conclude with proposed best-practices when using machine learning in autism research, and highlight some especially promising areas for collaborative work at the intersection of computational and behavioral science. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
United States | 5 | 42% |
India | 1 | 8% |
Brazil | 1 | 8% |
United Kingdom | 1 | 8% |
Canada | 1 | 8% |
Unknown | 3 | 25% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Members of the public | 6 | 50% |
Scientists | 5 | 42% |
Science communicators (journalists, bloggers, editors) | 1 | 8% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
United States | 3 | 1% |
Israel | 1 | <1% |
Mexico | 1 | <1% |
Canada | 1 | <1% |
Unknown | 286 | 98% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Ph. D. Student | 47 | 16% |
Researcher | 45 | 15% |
Student > Master | 27 | 9% |
Student > Bachelor | 25 | 9% |
Student > Doctoral Student | 20 | 7% |
Other | 53 | 18% |
Unknown | 75 | 26% |
Readers by discipline | Count | As % |
---|---|---|
Computer Science | 59 | 20% |
Psychology | 33 | 11% |
Engineering | 26 | 9% |
Medicine and Dentistry | 22 | 8% |
Neuroscience | 13 | 4% |
Other | 44 | 15% |
Unknown | 95 | 33% |