Reinforcement Learning for Personalised Health Informatics Interventions to Improve Patient Outcomes. ‘African Journal of Disability’ 2026 Special Collection
African Journal of Disability 2026 Special Collection: We invite you to submit
AOSIS calls on all authors to participate in the African Journal of Disability 2026 special collection that will be published in the open-access scholarly journal. Submit your latest research for consideration, contribute to the open-access content available to everyone, and share your expertise with a wider audience.
Timeline:
- Submissions open: 05 May 2025
- Submissions deadline: 05 July 2025
- Expected publication date: April 2026
Reinforcement Learning for Personalised Health Informatics Interventions to Improve Patient Outcomes
Utilising reinforcement learning (RL), personalised health informatics may customise interventions to each patient’s needs, potentially enhancing health outcomes. This technique is a useful tool in the era of personalised medicine since it can dynamically adapt to changing health circumstances, patient preferences, and behaviors. Real-time data allows RL to continuously learn and change therapies in response to changing patient behaviors and health circumstances. This enables optimisation of interventions to enhance adherence, engagement, and overall health outcomes. RL can improve methods for promoting diet, exercise, and medication compliance, among other healthy habits. Additionally, it can customise chronic illness treatment regimens by changing prescription dosages and establishing subsequent appointments. Furthermore, RL facilitates more efficient resource allocation for healthcare practitioners, ensuring prompt and appropriate decisions.
A value-based reinforcement learning algorithm called Q-Learning determines the value of actions in various situations. To handle complicated state spaces, Deep Q-Networks (DQN) integrate Q-Learning with deep neural networks. Actor-Critic Methods combine value-based and policy-based approaches for more effective learning, while Policy Gradient Methods learn the policy that maps states to actions directly, making them suitable in continuous action spaces. TensorFlow, PyTorch, OpenAI Gym, and RLlib are popular frameworks and libraries for implementing RL algorithms into action and evaluating them. Wearable technology integration with reinforcement learning enables continuous monitoring and modification. RL has the ability to offer personalised, real-time recommendations in telehealth applications. Additionally, it can be applied to customise mental health interventions such as CBT and meditation. Additionally, by continuously improving treatment programs, RL assists in the management of chronic illnesses like asthma, diabetes, and hypertension.
Collaborative, comprehensive patient care is made possible by the use of several RL agents in multi-agent systems. Patients and healthcare professionals build trust when RL models are more transparent. Further AI methods, including natural language processing (NLP), can be integrated with reinforcement learning (RL) to provide more complete health informatics solutions. Moreover, developing scalable reinforcement learning solutions ensures that they can handle big populations and varied patient data in an efficient manner. Data security and privacy can be ensured by methods like safe multi-party computation and differential privacy. Adoption and confidence are increased when RL models are made understandable to physicians and patients using techniques like feature significance and model distillation. Ensuring fairness in treatment suggestions and reducing biased decision-making are necessary for addressing ethical concerns. Furthermore, adversarial training and robust optimisation techniques to handle errors and noise in patient data can be used to improve the robustness and reliability of RL models.
Objective:
We believe this special collection will provide valuable insights and foster meaningful discussions among scholars and practitioners.
Recommended topics:
We encourage submissions on a range of topics, including but not limited to:
- Enhancing Health Outcomes through Personalized Interventions with Reinforcement Learning.
- Role of Real-Time Data in Reinforcement Learning for Dynamic Health Interventions.
- Optimising Patient Engagement and Adherence through RL-Based Personalized Health Informatics.
- Employing RL for Continuous Adaptation in Personalized Health Care.
- Integration of Reinforcement Learning with Health Informatics for Improved Patient Outcomes.
- Exploring the Effectiveness of Q-Learning in Personalized Health Informatics.
- Applications of RL Algorithms with TensorFlow, PyTorch, OpenAI Gym, and RLlib in Health Informatics.
- Contribution of Reinforcement Learning in Real-Time Health Data Analysis.
- Analysis of NLP and RL for Comprehensive Health Informatics Solutions.
- Advancements in Adoption of RL Models with Interpretability Techniques for Physicians and Patients.
Manuscript information:
The author guidelines include information about the types of articles received for publication and preparing a manuscript for submission. Read the full submissions guidelines.
Submission procedure:
When submitting your article to African Journal of Disability, choose ‘Reinforcement Learning for Personalised Health Informatics Interventions to Improve Patient Outcomes’ as the article type. You can access the submission portal on the journal’s website after logging in with your personal credentials. For further information on the submission process, visit the journal procedure page.
All submissions will undergo an anonymous review process to guarantee high scientific quality and relevance to the subject. The Editor-in-Chief will make the final decision on acceptance, revision, or rejection based on the feedback from the reviewers.
We will be happy to provide you with any assistance during the submission and application process. Kindly enquire at submissions@ajod.org.
All submissions and inquiries should be directed to the attention of the guest editors:
- Managing Guest Editor: Prof Thiago Goncalves dos Santos Martins (Federal University of Rio de Janeiro, Brazil) – thigonmart1224@yahoo.com
- First Co-Guest Editor: Prof Sunita Sahu (GD Goenka University, India) – sunita.sahu@ves.ac.in
- Second Co-Guest Editor: Mr. Sanjog Sigdel (Kathmandu University, Nepal) – sigdelsanjog@gmail.com
We would be honoured to receive a positive reply from you and look forward to receiving your article.
About AOSIS scholarly journals
Maximise journal impact through open access, expert publishing services, quality peer reviews, and cost-effective production. You can browse the AOSIS open-access journal collection per scholarly field and browse by keyword or per author at library.aosis.co.za. All our open-access content is free and available to everyone upon publication. A physical copy can be ordered at the AOSIS Shop.
Open access publishing
We are an open-access publisher, meaning all content is freely available without charge to the reader. Readers are allowed to read, download, copy, distribute, print, search, or link to the full texts of the books or use them for any other lawful purpose without asking prior permission from the publisher or the author, provided that the work is appropriately cited.