Forthcoming Articles

International Journal of Artificial Intelligence in Healthcare

International Journal of Artificial Intelligence in Healthcare (IJAIH)

Forthcoming articles have been peer-reviewed and accepted for publication but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.

Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.

Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.

Online First articles are also listed here. Online First articles are fully citeable, complete with a DOI. They can be cited, read, and downloaded. Online First articles are published as Open Access (OA) articles to make the latest research available as early as possible.

Open AccessArticles marked with this Open Access icon are Online First articles. They are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.

Register for our alerting service, which notifies you by email when new issues are published online.

(3 papers in press)

Regular Issues

  • Machine learning-based computer assisted model for visually impaired patients   Order a copy of this article
    by Manoranjan Dash 
    Abstract: A significant section of the population currently faces everyday obstacles due to blindness. This study presents a computer-assisted model (CAM) that uses machine learning to help visually impaired people become more independent and live better lives. The proposed model used natural language processing (NLP) to convert visual input into descriptive audio feedback and convolutional neural networks (CNNs) for real-time object detection and picture recognition. The system consists of an audio output interface, server-side processing, and a web application with camera access. The camera records live images, which are processed by machine learning algorithms to recognise text and objects and give the user audio descriptions. The purpose of the test is to determine whether the model can help visually impaired persons navigate and be more aware of their surroundings. The efficacy of the suggested methodology has been confirmed from the good accuracy (97.00%) and precision (96.50%).
    Keywords: machine learning; assistive technology; visually impaired; convolutional neural networks; CNNs; image recognition; natural language processing; NLP.
    DOI: 10.1504/IJAIH.2026.10080570
     
  • RR-interval based atrial fibrillation subtype classification using CALNet   Order a copy of this article
    by Sourov Das, Bijoy Das, Ainul Anam Shahjamal Khan 
    Abstract: An accurate classification of atrial fibrillation (AF) into its three paroxysmal, persistent, and permanent forms is important for medical treatment. Traditional diagnostic models rely on the use of binary discrimination of short ECG recordings. So, their limited application in practical situations. To address this, CALNet, a hybrid deep learning architecture, is proposed that can classify all three forms of AF at once. This model takes advantage of RR intervals obtained from 24-hour long-term ECG data in order to detect complex temporal patterns in the heart rhythm. Despite the difficulty in multi-class classification of continuous data, CALNet shows impressive results with an accuracy of 87.91%, AUC of 93.68%, and Specificity of 89.17%.
    Keywords: atrial fibrillation; AF; RR interval; deep learning model; AF subtypes; CALNet.
    DOI: 10.1504/IJAIH.2026.10081070
     
  • Multimodal EEG and clinical feature fusion for edge-deployable three-class dementia screening using lightweight GRU networks   Order a copy of this article
    by Seshan Saravanan, B. Shaanthapriyan, Gnanou Florence Sudha 
    Abstract: Differentiating Alzheimers disease (AD) and frontotemporal dementia (FTD) from cognitively normal (CN) individuals typically requires costly neuroimaging, and EEG-based deep learning methods are rarely evaluated under embedded deployment constraints. This study fuses resting-state EEG with clinical variables (age, gender, MMSE) for three-class dementia screening, systematically comparing single-layer GRU, two-layer GRU and CNN-GRU architectures to test whether added depth improves accuracy. The single-layer GRU matches deeper alternatives whilst cutting ARM Cortex-A53 latency by approximately 40%. Ten-fold subject-level cross-validation on 88 participants yielded 88.7 ? 2.4% accuracy (F1: 0.91), an 11-point gain over an EEG-only baseline. The INT8-quantised model sustains 87.7% accuracy at 26 ms inference on embedded ARM hardware, demonstrating clinically relevant accuracy without deep or cloud-dependent architectures and establishing a validated blueprint for point-of-care dementia screening.
    Keywords: Alzheimer’s disease; frontotemporal dementia; FTD; electroencephalography; EEG; edge deployment; ARM processor; model quantisation; dementia; embedded inference.
    DOI: 10.1504/IJAIH.2026.10081113
     

 

Return to top