Deep Learning to Detect Medical Treatment Fraud

Daniel Lasaga, Prakash Santhana
Proceedings of the KDD 2017: Workshop on Anomaly Detection in Finance, PMLR 71:114-120, 2018.

Abstract

Excessive treatment or testing of patients is considered one of the most ubiquitous and persistent forms of waste and abuse in healthcare. Some estimates show excessive treatment to be as high as 8% of all medical insurance provider expenditures. It is very difficult to identify an extraneous or unnecessary procedure or drug because there is such a wide variety of diagnoses and an equally large number of treatment options. Our goal in this paper was to show how RBMs can be utilized effectively to ferret out abnormal treatments where the prescribed treatment for a given diagnosis is not strictly followed. To test our hypothesis we generated 200,000 different injuries and injected 10% of the injuries with unnecessary treatments to reflect estimated industry prevalence levels. Using testing and training sets we found that Restricted Boltzmann Machines (RBMs) were able to reach AUCs of .95, lifts at 9.5 and recalls at 50%. Implementing our approach on real-world client datasets have shown performances levels that approach simulation performances despite additional noise.

Cite this Paper

BibTeX
@InProceedings{pmlr-v71-lasaga18a, title = {Deep Learning to Detect Medical Treatment Fraud}, author = {Lasaga, Daniel and Santhana, Prakash}, booktitle = {Proceedings of the KDD 2017: Workshop on Anomaly Detection in Finance}, pages = {114--120}, year = {2018}, editor = {Anandakrishnan, Archana and Kumar, Senthil and Statnikov, Alexander and Faruquie, Tanveer and Xu, Di}, volume = {71}, series = {Proceedings of Machine Learning Research}, month = {14 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v71/lasaga18a/lasaga18a.pdf}, url = {https://proceedings.mlr.press/v71/lasaga18a.html}, abstract = {Excessive treatment or testing of patients is considered one of the most ubiquitous and persistent forms of waste and abuse in healthcare. Some estimates show excessive treatment to be as high as 8% of all medical insurance provider expenditures. It is very difficult to identify an extraneous or unnecessary procedure or drug because there is such a wide variety of diagnoses and an equally large number of treatment options. Our goal in this paper was to show how RBMs can be utilized effectively to ferret out abnormal treatments where the prescribed treatment for a given diagnosis is not strictly followed. To test our hypothesis we generated 200,000 different injuries and injected 10% of the injuries with unnecessary treatments to reflect estimated industry prevalence levels. Using testing and training sets we found that Restricted Boltzmann Machines (RBMs) were able to reach AUCs of .95, lifts at 9.5 and recalls at 50%. Implementing our approach on real-world client datasets have shown performances levels that approach simulation performances despite additional noise.} }
Endnote
%0 Conference Paper %T Deep Learning to Detect Medical Treatment Fraud %A Daniel Lasaga %A Prakash Santhana %B Proceedings of the KDD 2017: Workshop on Anomaly Detection in Finance %C Proceedings of Machine Learning Research %D 2018 %E Archana Anandakrishnan %E Senthil Kumar %E Alexander Statnikov %E Tanveer Faruquie %E Di Xu %F pmlr-v71-lasaga18a %I PMLR %P 114--120 %U https://proceedings.mlr.press/v71/lasaga18a.html %V 71 %X Excessive treatment or testing of patients is considered one of the most ubiquitous and persistent forms of waste and abuse in healthcare. Some estimates show excessive treatment to be as high as 8% of all medical insurance provider expenditures. It is very difficult to identify an extraneous or unnecessary procedure or drug because there is such a wide variety of diagnoses and an equally large number of treatment options. Our goal in this paper was to show how RBMs can be utilized effectively to ferret out abnormal treatments where the prescribed treatment for a given diagnosis is not strictly followed. To test our hypothesis we generated 200,000 different injuries and injected 10% of the injuries with unnecessary treatments to reflect estimated industry prevalence levels. Using testing and training sets we found that Restricted Boltzmann Machines (RBMs) were able to reach AUCs of .95, lifts at 9.5 and recalls at 50%. Implementing our approach on real-world client datasets have shown performances levels that approach simulation performances despite additional noise.
APA
Lasaga, D. & Santhana, P.. (2018). Deep Learning to Detect Medical Treatment Fraud. Proceedings of the KDD 2017: Workshop on Anomaly Detection in Finance, in Proceedings of Machine Learning Research 71:114-120 Available from https://proceedings.mlr.press/v71/lasaga18a.html.

Related Material