Explainable unsupervised anomaly detection for healthcare insurance data
Abstract Background Waste and fraud are important problems for health insurers to deal with. With the advent of big data, these insurers are looking more and more towards data mining and machine learning methods to help in detecting waste and fraud. However, labeled data is costly and difficult to a...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-024-02823-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841544591406268416 |
|---|---|
| author | Hannes De Meulemeester Frank De Smet Johan van Dorst Elise Derroitte Bart De Moor |
| author_facet | Hannes De Meulemeester Frank De Smet Johan van Dorst Elise Derroitte Bart De Moor |
| author_sort | Hannes De Meulemeester |
| collection | DOAJ |
| description | Abstract Background Waste and fraud are important problems for health insurers to deal with. With the advent of big data, these insurers are looking more and more towards data mining and machine learning methods to help in detecting waste and fraud. However, labeled data is costly and difficult to acquire as it requires expert investigators and known care providers with atypical behavior. Methods In this work we show how recent advances in machine learning can be used to set up a workflow that can aid investigators in discovering practitioners or groups of practitioners with unusual resource use in order to more efficiently combat waste and fraud. We combine three different techniques, which have not been used in the context of healthcare insurance anomaly detection: categorical embeddings to deal with high-cardinality categorical variables, state-of-the-art unsupervised anomaly detection techniques to detect anomalies and Shapley additive explanations (SHAP) to explain the model output. Results The method has been evaluated on providers with a known anomalous profile and with the help of experts of the largest health insurance fund in Belgium. The quantitative experiments show that categorical embeddings offer a significant improvement compared to standard methods and that the state-of-the-art unsupervised anomaly detection techniques generally show an improvement over traditional methods. In a practical setting, the proposed workflow with SHAP was able to detect a previously unknown, anomalous trend among general practitioners. Conclusions The proposed workflow is able to detect known care providers with atypical behaviour and helps expert investigators in making informed decisions concerning possible fraud or overconsumption in the health insurance field. |
| format | Article |
| id | doaj-art-6cb8ec1b32a54b08ab227b7125670df8 |
| institution | Kabale University |
| issn | 1472-6947 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj-art-6cb8ec1b32a54b08ab227b7125670df82025-01-12T12:26:22ZengBMCBMC Medical Informatics and Decision Making1472-69472025-01-0125111110.1186/s12911-024-02823-6Explainable unsupervised anomaly detection for healthcare insurance dataHannes De Meulemeester0Frank De Smet1Johan van Dorst2Elise Derroitte3Bart De Moor4Department of Electrical Engineering, ESAT-STADIUS, KU LeuvenChristian Health Insurance FundChristian Health Insurance FundChristian Health Insurance FundDepartment of Electrical Engineering, ESAT-STADIUS, KU LeuvenAbstract Background Waste and fraud are important problems for health insurers to deal with. With the advent of big data, these insurers are looking more and more towards data mining and machine learning methods to help in detecting waste and fraud. However, labeled data is costly and difficult to acquire as it requires expert investigators and known care providers with atypical behavior. Methods In this work we show how recent advances in machine learning can be used to set up a workflow that can aid investigators in discovering practitioners or groups of practitioners with unusual resource use in order to more efficiently combat waste and fraud. We combine three different techniques, which have not been used in the context of healthcare insurance anomaly detection: categorical embeddings to deal with high-cardinality categorical variables, state-of-the-art unsupervised anomaly detection techniques to detect anomalies and Shapley additive explanations (SHAP) to explain the model output. Results The method has been evaluated on providers with a known anomalous profile and with the help of experts of the largest health insurance fund in Belgium. The quantitative experiments show that categorical embeddings offer a significant improvement compared to standard methods and that the state-of-the-art unsupervised anomaly detection techniques generally show an improvement over traditional methods. In a practical setting, the proposed workflow with SHAP was able to detect a previously unknown, anomalous trend among general practitioners. Conclusions The proposed workflow is able to detect known care providers with atypical behaviour and helps expert investigators in making informed decisions concerning possible fraud or overconsumption in the health insurance field.https://doi.org/10.1186/s12911-024-02823-6Health insuranceAnomaly detectionUnsupervised machine learning |
| spellingShingle | Hannes De Meulemeester Frank De Smet Johan van Dorst Elise Derroitte Bart De Moor Explainable unsupervised anomaly detection for healthcare insurance data BMC Medical Informatics and Decision Making Health insurance Anomaly detection Unsupervised machine learning |
| title | Explainable unsupervised anomaly detection for healthcare insurance data |
| title_full | Explainable unsupervised anomaly detection for healthcare insurance data |
| title_fullStr | Explainable unsupervised anomaly detection for healthcare insurance data |
| title_full_unstemmed | Explainable unsupervised anomaly detection for healthcare insurance data |
| title_short | Explainable unsupervised anomaly detection for healthcare insurance data |
| title_sort | explainable unsupervised anomaly detection for healthcare insurance data |
| topic | Health insurance Anomaly detection Unsupervised machine learning |
| url | https://doi.org/10.1186/s12911-024-02823-6 |
| work_keys_str_mv | AT hannesdemeulemeester explainableunsupervisedanomalydetectionforhealthcareinsurancedata AT frankdesmet explainableunsupervisedanomalydetectionforhealthcareinsurancedata AT johanvandorst explainableunsupervisedanomalydetectionforhealthcareinsurancedata AT elisederroitte explainableunsupervisedanomalydetectionforhealthcareinsurancedata AT bartdemoor explainableunsupervisedanomalydetectionforhealthcareinsurancedata |