Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models
IntroductionIndividuals in the midst of a mental health crisis frequently exhibit instability and face an elevated risk of recurring crises in the subsequent weeks, which underscores the importance of timely intervention in mental healthcare. This work presents a data-driven method to infer the ment...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Digital Health |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fdgth.2024.1322555/full |
| _version_ | 1797334658362376192 |
|---|---|
| author | Roger Garriga Roger Garriga Vicenç Gómez Gábor Lugosi Gábor Lugosi Gábor Lugosi |
| author_facet | Roger Garriga Roger Garriga Vicenç Gómez Gábor Lugosi Gábor Lugosi Gábor Lugosi |
| author_sort | Roger Garriga |
| collection | DOAJ |
| description | IntroductionIndividuals in the midst of a mental health crisis frequently exhibit instability and face an elevated risk of recurring crises in the subsequent weeks, which underscores the importance of timely intervention in mental healthcare. This work presents a data-driven method to infer the mental state of a patient during the weeks following a mental health crisis by leveraging their historical data. Additionally, we propose a policy that determines the necessary duration for closely monitoring a patient after a mental health crisis before considering them stable.MethodsWe model the patient’s mental state as a Hidden Markov Process, partially observed through mental health crisis events. We introduce a closed-form solution that leverages the model parameters to optimally estimate the risk of future mental health crises. Our policy determines a patient should be closely monitored when their estimated risk of crisis exceeds a predefined threshold. The method’s performance is evaluated using both simulated data and a real-world dataset comprising 162 anonymized psychiatric patients.ResultsIn the simulations, 96.2% of the patients identified by the policy were in an unstable state, achieving a F1 score of 0.74. In the real-world dataset, the policy yielded an F1 score of 0.79, with a sensitivity of 79.8% and specificity of 88.9%. Under this policy, 67.3% of the patients should undergo close monitoring for one week, 21.6% during 2 weeks or more, while 11.1% do not need close monitoring.DiscussionThe simulation results provide compelling evidence that the method is effective under the specified assumptions. When applied to actual psychiatric patients, the proposed policy showed significant potential for providing an individualized assessment of the required duration for close and automatic monitoring after a mental health crisis to reduce the relapse risks. |
| first_indexed | 2024-03-08T08:24:17Z |
| format | Article |
| id | doaj.art-42ca815ed9fb401e991a72fd83d366ee |
| institution | Directory Open Access Journal |
| issn | 2673-253X |
| language | English |
| last_indexed | 2024-03-08T08:24:17Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Digital Health |
| spelling | doaj.art-42ca815ed9fb401e991a72fd83d366ee2024-02-02T04:52:28ZengFrontiers Media S.A.Frontiers in Digital Health2673-253X2024-02-01610.3389/fdgth.2024.13225551322555Individualized post-crisis monitoring of psychiatric patients via Hidden Markov modelsRoger Garriga0Roger Garriga1Vicenç Gómez2Gábor Lugosi3Gábor Lugosi4Gábor Lugosi5Koa Health, Barcelona, SpainDepartment of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, SpainDepartment of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, SpainICREA, Barcelona, SpainDepartment of Economics and Business, Universitat Pompeu Fabra, Barcelona, SpainBarcelona School of Economics, Barcelona, SpainIntroductionIndividuals in the midst of a mental health crisis frequently exhibit instability and face an elevated risk of recurring crises in the subsequent weeks, which underscores the importance of timely intervention in mental healthcare. This work presents a data-driven method to infer the mental state of a patient during the weeks following a mental health crisis by leveraging their historical data. Additionally, we propose a policy that determines the necessary duration for closely monitoring a patient after a mental health crisis before considering them stable.MethodsWe model the patient’s mental state as a Hidden Markov Process, partially observed through mental health crisis events. We introduce a closed-form solution that leverages the model parameters to optimally estimate the risk of future mental health crises. Our policy determines a patient should be closely monitored when their estimated risk of crisis exceeds a predefined threshold. The method’s performance is evaluated using both simulated data and a real-world dataset comprising 162 anonymized psychiatric patients.ResultsIn the simulations, 96.2% of the patients identified by the policy were in an unstable state, achieving a F1 score of 0.74. In the real-world dataset, the policy yielded an F1 score of 0.79, with a sensitivity of 79.8% and specificity of 88.9%. Under this policy, 67.3% of the patients should undergo close monitoring for one week, 21.6% during 2 weeks or more, while 11.1% do not need close monitoring.DiscussionThe simulation results provide compelling evidence that the method is effective under the specified assumptions. When applied to actual psychiatric patients, the proposed policy showed significant potential for providing an individualized assessment of the required duration for close and automatic monitoring after a mental health crisis to reduce the relapse risks.https://www.frontiersin.org/articles/10.3389/fdgth.2024.1322555/fullprobabilistic modelingmental health crisisHidden Markov modelmental healthpsychiatrymachine learning |
| spellingShingle | Roger Garriga Roger Garriga Vicenç Gómez Gábor Lugosi Gábor Lugosi Gábor Lugosi Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models Frontiers in Digital Health probabilistic modeling mental health crisis Hidden Markov model mental health psychiatry machine learning |
| title | Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models |
| title_full | Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models |
| title_fullStr | Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models |
| title_full_unstemmed | Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models |
| title_short | Individualized post-crisis monitoring of psychiatric patients via Hidden Markov models |
| title_sort | individualized post crisis monitoring of psychiatric patients via hidden markov models |
| topic | probabilistic modeling mental health crisis Hidden Markov model mental health psychiatry machine learning |
| url | https://www.frontiersin.org/articles/10.3389/fdgth.2024.1322555/full |
| work_keys_str_mv | AT rogergarriga individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels AT rogergarriga individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels AT vicencgomez individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels AT gaborlugosi individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels AT gaborlugosi individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels AT gaborlugosi individualizedpostcrisismonitoringofpsychiatricpatientsviahiddenmarkovmodels |