Artificial Intelligence-Driven Digital Phenotyping in Psychiatry: Clinical Utility, Ethical Challenges, and Future Directions
Fuente:
Zenodo
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Veröffentlicht: |
Zenodo
2026
|
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866901829177049088 |
|---|---|
| author | Anna Byr, Melania Majewska, Natalia Piasecka, Izabela Rafalska, Jakub Smagoń, Natalia Kornacka, Martyna Kaim, Joanna Bober, Magdalena Bochenek and Wiktoria Siewiera |
| author_facet | Anna Byr, Melania Majewska, Natalia Piasecka, Izabela Rafalska, Jakub Smagoń, Natalia Kornacka, Martyna Kaim, Joanna Bober, Magdalena Bochenek and Wiktoria Siewiera |
| contents | <p>Digital phenotyping, defined as the moment-by-moment quantification of individual-level human phenotype using data from <br>personal digital devices, has emerged as a promising approach to psychiatric assessment and monitoring. When combined with artificial intelligence (AI) and machine learning algorithms, digital phenotyping enables passive, continuous, and ecologically valid measurement of behavioral and cognitive markers relevant to psychiatric disorders. This interdisciplinary review examines the current state of AI-driven digital phenotyping in psychiatry, with emphasis on its clinical applications across major psychiatric conditions including depression, schizophrenia, bipolar disorder, anxiety disorders, and suicide risk. We discuss the methodological landscape — spanning passive sensing, natural language processing, and multimodal data fusion — and critically evaluate evidence regarding the predictive validity and clinical utility of these approaches. Furthermore, we address the significant ethical, legal, and social challenges inherent in deploying AI-based monitoring technologies in vulnerable psychiatric populations, including issues of informed consent, data privacy, algorithmic bias, and therapeutic relationship. Finally, we outline future directions for the field, including the integration of federated learning, personalized medicine frameworks, and standardized outcome metrics. We conclude that while AI-driven digital phenotyping holds transformative potential for psychiatry, its responsible clinical implementation requires robust interdisciplinary governance, equitable algorithm design, and meaningful patient engagement. </p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19702516 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Artificial Intelligence-Driven Digital Phenotyping in Psychiatry: Clinical Utility, Ethical Challenges, and Future Directions Anna Byr, Melania Majewska, Natalia Piasecka, Izabela Rafalska, Jakub Smagoń, Natalia Kornacka, Martyna Kaim, Joanna Bober, Magdalena Bochenek and Wiktoria Siewiera <p>Digital phenotyping, defined as the moment-by-moment quantification of individual-level human phenotype using data from <br>personal digital devices, has emerged as a promising approach to psychiatric assessment and monitoring. When combined with artificial intelligence (AI) and machine learning algorithms, digital phenotyping enables passive, continuous, and ecologically valid measurement of behavioral and cognitive markers relevant to psychiatric disorders. This interdisciplinary review examines the current state of AI-driven digital phenotyping in psychiatry, with emphasis on its clinical applications across major psychiatric conditions including depression, schizophrenia, bipolar disorder, anxiety disorders, and suicide risk. We discuss the methodological landscape — spanning passive sensing, natural language processing, and multimodal data fusion — and critically evaluate evidence regarding the predictive validity and clinical utility of these approaches. Furthermore, we address the significant ethical, legal, and social challenges inherent in deploying AI-based monitoring technologies in vulnerable psychiatric populations, including issues of informed consent, data privacy, algorithmic bias, and therapeutic relationship. Finally, we outline future directions for the field, including the integration of federated learning, personalized medicine frameworks, and standardized outcome metrics. We conclude that while AI-driven digital phenotyping holds transformative potential for psychiatry, its responsible clinical implementation requires robust interdisciplinary governance, equitable algorithm design, and meaningful patient engagement. </p> |
| title | Artificial Intelligence-Driven Digital Phenotyping in Psychiatry: Clinical Utility, Ethical Challenges, and Future Directions |
| url | https://doi.org/10.5281/zenodo.19702516 |