Academic Journal
Deep reinforcement learning for Type 1 Diabetes: Dual PPO controller for personalized insulin management
| Τίτλος: | Deep reinforcement learning for Type 1 Diabetes: Dual PPO controller for personalized insulin management |
|---|---|
| Συγγραφείς: | Marchetti A., Sasso D., D'Antoni F., Morandin F., Parton M., Matarrese M. A. G., Merone M. |
| Συνεισφορές: | Marchetti, A., Sasso, D., D'Antoni, F., Morandin, F., Parton, M., Matarrese, M. A. G., Merone, M. |
| Έτος έκδοσης: | 2025 |
| Συλλογή: | ARPI - Archivio della Ricerca dell'Università di Pisa |
| Θεματικοί όροι: | Diabete, Glycemic control, Precision medicine, Proximal Policy Optimization, Reinforcement learning |
| Περιγραφή: | Background: Managing blood glucose levels in Type 1 Diabetes Mellitus (T1DM) is essential to prevent complications. Traditional insulin delivery methods often require significant patient involvement, limiting automation. Reinforcement Learning (RL)-based controllers offer a promising approach for improving automated insulin administration. Methods: We propose a Dual Proximal Policy Optimization (Dual PPO) controller for personalized insulin delivery in a hybrid closed-loop system. The controller optimizes patient-specific insulin bounds through a grid search on pre-trained models to manage both hyperglycemia and hypoglycemia. A safe-control mechanism prevents insulin administration when glucose levels drop below a predefined threshold. The system was evaluated on 10 in silico adult patients using the UVA/Padova simulator, with five-day randomized meal scenarios. Results: The Dual PPO controller significantly improved Time in Range (TIR) (69.30% ±1.61) compared to a single PPO model (61.69% ±1.54). The system effectively reduced severe hyperglycemia while maintaining a low incidence of severe hypoglycemia. Unlike conventional open-loop methods such as Basal-Bolus (BBC) and Proportional–Integral–Derivative (PIDC) controllers, our system requires minimal patient interaction, eliminating the need for carbohydrate estimation. Conclusions: The Dual PPO controller enhances personalized insulin delivery in T1DM, improving glycemic control while reducing patient burden. This approach advances precision medicine in diabetes management, with potential for future real-world applications. |
| Τύπος εγγράφου: | article in journal/newspaper |
| Γλώσσα: | English |
| Relation: | info:eu-repo/semantics/altIdentifier/pmid/40239234; volume:191; numberofpages:13; journal:COMPUTERS IN BIOLOGY AND MEDICINE; https://hdl.handle.net/11568/1343693; https://www.sciencedirect.com/science/article/pii/S0010482525004986?via=ihub |
| DOI: | 10.1016/j.compbiomed.2025.110147 |
| Διαθεσιμότητα: | https://hdl.handle.net/11568/1343693 https://doi.org/10.1016/j.compbiomed.2025.110147 https://www.sciencedirect.com/science/article/pii/S0010482525004986?via=ihub |
| Αριθμός Καταχώρησης: | edsbas.12B1DB49 |
| Βάση Δεδομένων: | BASE |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://hdl.handle.net/11568/1343693# Name: EDS - BASE (ns324271) Category: fullText Text: View record from BASE |
|---|---|
| Header | DbId: edsbas DbLabel: BASE An: edsbas.12B1DB49 RelevancyScore: 900 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 899.935607910156 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Deep reinforcement learning for Type 1 Diabetes: Dual PPO controller for personalized insulin management – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Marchetti+A%2E%22">Marchetti A.</searchLink><br /><searchLink fieldCode="AR" term="%22Sasso+D%2E%22">Sasso D.</searchLink><br /><searchLink fieldCode="AR" term="%22D'Antoni+F%2E%22">D'Antoni F.</searchLink><br /><searchLink fieldCode="AR" term="%22Morandin+F%2E%22">Morandin F.</searchLink><br /><searchLink fieldCode="AR" term="%22Parton+M%2E%22">Parton M.</searchLink><br /><searchLink fieldCode="AR" term="%22Matarrese+M%2E+A%2E+G%2E%22">Matarrese M. A. G.</searchLink><br /><searchLink fieldCode="AR" term="%22Merone+M%2E%22">Merone M.</searchLink> – Name: Author Label: Contributors Group: Au Data: Marchetti, A.<br />Sasso, D.<br />D'Antoni, F.<br />Morandin, F.<br />Parton, M.<br />Matarrese, M. A. G.<br />Merone, M. – Name: DatePubCY Label: Publication Year Group: Date Data: 2025 – Name: Subset Label: Collection Group: HoldingsInfo Data: ARPI - Archivio della Ricerca dell'Università di Pisa – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Diabete%22">Diabete</searchLink><br /><searchLink fieldCode="DE" term="%22Glycemic+control%22">Glycemic control</searchLink><br /><searchLink fieldCode="DE" term="%22Precision+medicine%22">Precision medicine</searchLink><br /><searchLink fieldCode="DE" term="%22Proximal+Policy+Optimization%22">Proximal Policy Optimization</searchLink><br /><searchLink fieldCode="DE" term="%22Reinforcement+learning%22">Reinforcement learning</searchLink> – Name: Abstract Label: Description Group: Ab Data: Background: Managing blood glucose levels in Type 1 Diabetes Mellitus (T1DM) is essential to prevent complications. Traditional insulin delivery methods often require significant patient involvement, limiting automation. Reinforcement Learning (RL)-based controllers offer a promising approach for improving automated insulin administration. Methods: We propose a Dual Proximal Policy Optimization (Dual PPO) controller for personalized insulin delivery in a hybrid closed-loop system. The controller optimizes patient-specific insulin bounds through a grid search on pre-trained models to manage both hyperglycemia and hypoglycemia. A safe-control mechanism prevents insulin administration when glucose levels drop below a predefined threshold. The system was evaluated on 10 in silico adult patients using the UVA/Padova simulator, with five-day randomized meal scenarios. Results: The Dual PPO controller significantly improved Time in Range (TIR) (69.30% ±1.61) compared to a single PPO model (61.69% ±1.54). The system effectively reduced severe hyperglycemia while maintaining a low incidence of severe hypoglycemia. Unlike conventional open-loop methods such as Basal-Bolus (BBC) and Proportional–Integral–Derivative (PIDC) controllers, our system requires minimal patient interaction, eliminating the need for carbohydrate estimation. Conclusions: The Dual PPO controller enhances personalized insulin delivery in T1DM, improving glycemic control while reducing patient burden. This approach advances precision medicine in diabetes management, with potential for future real-world applications. – Name: TypeDocument Label: Document Type Group: TypDoc Data: article in journal/newspaper – Name: Language Label: Language Group: Lang Data: English – Name: NoteTitleSource Label: Relation Group: SrcInfo Data: info:eu-repo/semantics/altIdentifier/pmid/40239234; volume:191; numberofpages:13; journal:COMPUTERS IN BIOLOGY AND MEDICINE; https://hdl.handle.net/11568/1343693; https://www.sciencedirect.com/science/article/pii/S0010482525004986?via=ihub – Name: DOI Label: DOI Group: ID Data: 10.1016/j.compbiomed.2025.110147 – Name: URL Label: Availability Group: URL Data: https://hdl.handle.net/11568/1343693<br />https://doi.org/10.1016/j.compbiomed.2025.110147<br />https://www.sciencedirect.com/science/article/pii/S0010482525004986?via=ihub – Name: AN Label: Accession Number Group: ID Data: edsbas.12B1DB49 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.12B1DB49 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1016/j.compbiomed.2025.110147 Languages: – Text: English Subjects: – SubjectFull: Diabete Type: general – SubjectFull: Glycemic control Type: general – SubjectFull: Precision medicine Type: general – SubjectFull: Proximal Policy Optimization Type: general – SubjectFull: Reinforcement learning Type: general Titles: – TitleFull: Deep reinforcement learning for Type 1 Diabetes: Dual PPO controller for personalized insulin management Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Marchetti A. – PersonEntity: Name: NameFull: Sasso D. – PersonEntity: Name: NameFull: D'Antoni F. – PersonEntity: Name: NameFull: Morandin F. – PersonEntity: Name: NameFull: Parton M. – PersonEntity: Name: NameFull: Matarrese M. A. G. – PersonEntity: Name: NameFull: Merone M. – PersonEntity: Name: NameFull: Marchetti, A. – PersonEntity: Name: NameFull: Sasso, D. – PersonEntity: Name: NameFull: D'Antoni, F. – PersonEntity: Name: NameFull: Morandin, F. – PersonEntity: Name: NameFull: Parton, M. – PersonEntity: Name: NameFull: Matarrese, M. A. G. – PersonEntity: Name: NameFull: Merone, M. IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2025 Identifiers: – Type: issn-locals Value: edsbas |
| ResultId | 1 |