Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning
Dynamic Treatment Regimes (DTRs) are sets of sequential decision rules that can be adapted over time to treat patients with a specific pathology. DTR consists of alternative treatment paths and any of these treatments can be adapted depending on the patient's characteristics. Reinforcement Lear...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9837927/ |
| _version_ | 1811344615581679616 |
|---|---|
| author | Syed Ihtesham Hussain Shah Antonio Coronato Muddasar Naeem Giuseppe De Pietro |
| author_facet | Syed Ihtesham Hussain Shah Antonio Coronato Muddasar Naeem Giuseppe De Pietro |
| author_sort | Syed Ihtesham Hussain Shah |
| collection | DOAJ |
| description | Dynamic Treatment Regimes (DTRs) are sets of sequential decision rules that can be adapted over time to treat patients with a specific pathology. DTR consists of alternative treatment paths and any of these treatments can be adapted depending on the patient's characteristics. Reinforcement Learning (RL) and Imitation Learning (IL) approaches have been deployed for obtaining optimal treatment for a patient but, these approaches rely only on positive trajectories (i.e., treatments that concluded with positive responses of the patient). In contrast, negative trajectories (i.e., samples of non-responding treatments) are discarded, although these have valuable information content. We propose a Cooperative Imitation Learning (CIL) method that exploits information from both negative and positive trajectories to learn the optimal DTR. The proposed method reduces the chance of selecting any treatment which results in a negative outcome (negative response of the patient) during the medical examination. To validate our approach, we have considered a well-known DTR which is defined for the treatment of patients with alcohol addiction. Results show that our approach outperforms those that rely only on positive trajectories. |
| first_indexed | 2024-04-13T19:49:50Z |
| format | Article |
| id | doaj.art-2f40d1c4726f4744888b35b48f68d6ad |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-13T19:49:50Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-2f40d1c4726f4744888b35b48f68d6ad2022-12-22T02:32:34ZengIEEEIEEE Access2169-35362022-01-0110781487815810.1109/ACCESS.2022.31934949837927Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation LearningSyed Ihtesham Hussain Shah0https://orcid.org/0000-0002-6390-1864Antonio Coronato1https://orcid.org/0000-0001-8177-032XMuddasar Naeem2https://orcid.org/0000-0003-0815-4883Giuseppe De Pietro3CNR, Institute for High Performance Computing and Networking (ICAR), Napoli, ItalyCNR, Institute for High Performance Computing and Networking (ICAR), Napoli, ItalyCNR, Institute for High Performance Computing and Networking (ICAR), Napoli, ItalyCNR, Institute for High Performance Computing and Networking (ICAR), Napoli, ItalyDynamic Treatment Regimes (DTRs) are sets of sequential decision rules that can be adapted over time to treat patients with a specific pathology. DTR consists of alternative treatment paths and any of these treatments can be adapted depending on the patient's characteristics. Reinforcement Learning (RL) and Imitation Learning (IL) approaches have been deployed for obtaining optimal treatment for a patient but, these approaches rely only on positive trajectories (i.e., treatments that concluded with positive responses of the patient). In contrast, negative trajectories (i.e., samples of non-responding treatments) are discarded, although these have valuable information content. We propose a Cooperative Imitation Learning (CIL) method that exploits information from both negative and positive trajectories to learn the optimal DTR. The proposed method reduces the chance of selecting any treatment which results in a negative outcome (negative response of the patient) during the medical examination. To validate our approach, we have considered a well-known DTR which is defined for the treatment of patients with alcohol addiction. Results show that our approach outperforms those that rely only on positive trajectories.https://ieeexplore.ieee.org/document/9837927/Inverse reinforcement learningimitation learningdynamic treatment regimereinforcement learningcooperative imitation learningMarkov decision process |
| spellingShingle | Syed Ihtesham Hussain Shah Antonio Coronato Muddasar Naeem Giuseppe De Pietro Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning IEEE Access Inverse reinforcement learning imitation learning dynamic treatment regime reinforcement learning cooperative imitation learning Markov decision process |
| title | Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning |
| title_full | Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning |
| title_fullStr | Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning |
| title_full_unstemmed | Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning |
| title_short | Learning and Assessing Optimal Dynamic Treatment Regimes Through Cooperative Imitation Learning |
| title_sort | learning and assessing optimal dynamic treatment regimes through cooperative imitation learning |
| topic | Inverse reinforcement learning imitation learning dynamic treatment regime reinforcement learning cooperative imitation learning Markov decision process |
| url | https://ieeexplore.ieee.org/document/9837927/ |
| work_keys_str_mv | AT syedihteshamhussainshah learningandassessingoptimaldynamictreatmentregimesthroughcooperativeimitationlearning AT antoniocoronato learningandassessingoptimaldynamictreatmentregimesthroughcooperativeimitationlearning AT muddasarnaeem learningandassessingoptimaldynamictreatmentregimesthroughcooperativeimitationlearning AT giuseppedepietro learningandassessingoptimaldynamictreatmentregimesthroughcooperativeimitationlearning |