Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 2 appraisals
PloS one
A hybrid deep learning model for user story effort estimation
Accurate effort estimation of user stories is a key challenge in agile software development due to both subjectivity and the complexity of natural language requirements. This paper proposes a hybrid Deep Learning (DL) model for data driven effort estimation using large scale textual data and advanced semantic modeling. One of the significant contributions is the development of a dataset of 6,956 user stories which was collected from many heterogeneous sources and then meticulously cleaned and refined to 4,079 high quality instances, by using a systematic preprocessing and expert validation. Following a Design Science Research (DSR) methodology, a hybrid model integrating a pre-trained Bidirectional Encoder Representations from Transformers (BERT) encoder with a Long Short-Term Memory (LSTM) is developed to capture both contextual semantics and sequential dependencies in user story description. The DL model is evaluated against multiple Machine Learning (ML) baselines using a robust multi-metric framework. Experimental findings show the superior performance of the proposed model with a Mean Absolute Error (MAE) = 0.6481, Root Mean Square Error (RMSE) = 1.4559 and = 0.6581, which is a huge improvement over the conventional methods. To ensure practical relevance in discrete Scrum planning, the continuous model outputs were mapped to the standard Fibonacci sequence, achieving a classification accuracy of 72%. To guarantee the consistency of performance improvements, the statistical validation is done with the Wilcoxon Signed-Rank Test to show that the improvements are significant (p < 0.05). Moreover, the model is operationalized as web based decision support system to enable real time estimation in agile software development. The results reveal the efficacy of using large scale curated datasets alongside hybrid DL model to reduce estimation bias and boost predictive quality to deliver a scalable and strong solution for intelligent agile project management. Future work will focus on integrating eXplainable AI (XAI) techniques and validating the model across real world industrial datasets to further enhance transparency and generalizability.
18 July 2026
Read appraisal →Artificial intelligence in medicine
CL-MHAD: Contrastive Learning-based Multi-Hypergraph Aggregation and Diffusion model for prescription recommendation
Multiple syndrome-based prescription recommendations are significant for personalized diagnosis and treatment in Traditional Chinese Medicine (TCM). However, it remains a challenge to effectively extract and fuse multi-dimensional knowledge in herbs for accurate syndrome-based prescription recommendations. To address it, we propose CL-MHAD, a contrastive learning-based multi-hypergraph aggregation and diffusion model for prescription recommendation. Focusing on prescriptions, herbs, properties, and dosages, this model presents a multi-view hypergraph reconstruction mechanism, which employs a consistent Top-K neighbor strategy to generate affinity and feature matrices. Subsequently, a diffusion-enhanced method is proposed, utilizing multi-step propagation and weighted mapping to capture high-order relationships. Then, we propose a cross-view contrastive learning strategy to enhance discrimination and present a topology-aware random walk augmentation approach to alleviate data sparsity. An integrated loss function combining classification, contrastive, and augmentation losses is designed to optimize model performance. We evaluate CL-MHAD in a real-world clinical setting using Gastrointestinal Diseases (GID) as a case study. Extensive experiments, including comparative evaluations, ablation studies, and parameter sensitivity analyses, show that CL-MHAD achieves overall superior performance compared with representative baselines across multiple evaluation metrics. Specifically, the model achieves performance gains of 1.27% to 24.67% over representative baselines across all datasets. We further evaluate the model through dosage-syndrome interaction and data-sparsity robustness tests, with results demonstrating the effectiveness of our weighted fusion strategy and the model's superior stability. This study offers an effective solution and a promising paradigm for accurate syndrome-based prescription recommendations in TCM.
4 May 2026
Read appraisal →