Dinithi Vithanage

Dinithi Vithanage

Auto-generated profile, based on publicly available information

Dinithi Vithanage

School of Computing and Information Technology, University of Wollongong, Wollongong, Australia

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 January 2024 in Journal of Healthcare Informatics Research
Verified
Journal Article
Closed access
PDF file
DVDinithi Vithanage
Dinithi Vithanage
Ping YuPing Yu
Ping Yu
LWLei Wang
Lei Wang
CDChao Deng
Chao Deng

20 February 2025 in Journal of Healthcare Informatics Research
Verified
Journal Article
Open access
PDF file
DVDinithi Vithanage
Dinithi Vithanage
CDChao Deng
Chao Deng
LWLei Wang
Lei Wang
MYMengyang Yin
Mengyang Yin
MAMohammad Alkhalaf
Mohammad Alkhalaf
Ping YuPing Yu
Ping Yu
Information extraction (IE) of unstructured electronic health records is challenging due to the semantic complexity of textual data. Generative large language models (LLMs) offer promising solutions to address this challenge. However, identifying the best training methods to adapt LLMs for IE in residential aged care settings remains underexplored. This research addresses this challenge by evaluating the effects of zero-shot and few-shot learning, both with and without parameter-efficient fine-tuning (PEFT) and retrieval-augmented generation (RAG) using Llama 3.1-8B. The study performed named entity recognition (NER) to nursing notes from Australian aged care facilities (RACFs), focusing on agitation in dementia and malnutrition risk factors. Performance evaluation includes accuracy, macro-averaged precision, recall, and F1 score. We used non-parametric statistical methods to compare if the differences were statistically significant. Results show that zero-shot and few-shot learning, whether combined with PEFT or RAG, achieve comparable performance across the clinical domains when the same prompting template is used. Few-shot learning significantly outperforms zero-shot learning when neither PEFT nor RAG is applied. Notably, PEFT significantly improves model performance in both zero-shot and few-shot learning; however, RAG significantly improves performance only in few-shot learning. After PEFT, the performance of zero-shot learning reaches a comparable level with few-shot learning. However, few-shot learning with RAG significantly outperforms zero-shot learning with RAG. We also found a similar level of performance between few-shot learning with RAG and zero-shot learning with PEFT. These findings provide valuable insights for researchers, practitioners, and stakeholders to optimize the use of generative LLMs in clinical IE.

30 August 2025 in Computers in Biology and Medicine
Verified
Journal Article
Open access
DVDinithi Vithanage
Dinithi Vithanage
Ping YuPing Yu
Ping Yu
QXQianqian Xie
Qianqian Xie
HXHua Xu
Hua Xu
LWLei Wang
Lei Wang
CDChao Deng
Chao Deng
Despite rapid healthcare digitization, extracting information from unstructured electronic health records (EHRs), such as nursing notes, remains challenging due to inconsistencies and ambiguities in clinical documentation. Generative large language models (LLMs) have emerged as promising tools for automating information extraction (IE); however, their application in real-world clinical settings, such as residential aged care (RAC), is limited by critical gaps. Prior studies have often focused on structured EHR data and conventional evaluation metrics such as accuracy and F1 score, overlooking critical aspects like robustness, fairness, bias, and contextual relevance, particularly in unstructured clinical narratives. To address these gaps, this study develops a holistic evaluation framework for clinical IE from free-text nursing notes in the Australian RAC. We systematically evaluate 17 LLMs, including general-purpose and healthcare-specific variants (e.g., LLaMA, Mistral, Gemini, T5) across retrieval-augmented generation (RAG) frameworks and few-shot learning configurations (one-shot, three-shot, four-shot, five-shot). The evaluation focuses on two clinical IE tasks: named entity recognition (NER) and summarization. Results reveal LLaMA 3.1 achieved 88.58 % accuracy, 87.43 % F1 score in NER, 88.18 % F1 score, and 83.15 % relevance in summarization. However, robustness remained low (4.00 % for NER, 4.31 % for summarization) despite excellent fairness (99.9 %) and minimal bias (0.11 %) in both tasks. Further, healthcare-specific LLMs slightly outperform general models, and RAG-based approaches (LangChain, LlamaIndex) yield superior results. Task-specific optimal few-shot settings emerged: three-shot for NER and five-shot for summarization. This study provides a foundation for safely integrating generative AI into clinical decision support.

25 January 2024
Verified
Book Chapter
Open access
PDF file
DVDinithi Vithanage
Dinithi Vithanage
YZYunshu Zhu
Yunshu Zhu
ZZZhenyu Zhang
Zhenyu Zhang
CDChao Deng
Chao Deng
MYMengyang Yin
Mengyang Yin
Ping YuPing Yu
Ping Yu
Nursing staff record observations about older people under their care in free-text nursing notes. These notes contain older people’s care needs, disease symptoms, frequency of symptom occurrence, nursing actions, etc. Therefore, it is vital to develop a technique to uncover important data from these notes. This study developed and evaluated a deep learning and transfer learning-based named entity recognition (NER) model for extracting symptoms of agitation in dementia from the nursing notes. We employed a Clinical BioBERT model for word embedding. Then we applied bidirectional long-short-term memory (BiLSTM) and conditional random field (CRF) models for NER on nursing notes from Australian residential aged care facilities. The proposed NER model achieves satisfactory performance in extracting symptoms of agitation in dementia with a 75% F1 score and 78% accuracy. We will further develop machine learning models to recommend the optimal nursing actions to manage agitation.

7 August 2025
Verified
Book Chapter
Open access
HZHengyi Zhang
Hengyi Zhang
DVDinithi Vithanage
Dinithi Vithanage
TSTing Song
Ting Song
CDChao Deng
Chao Deng
Ping YuPing Yu
Ping Yu
Unstructured electronic health records are a rich source of patient-specific information but are challenging for analysis due to inconsistent terminology, diverse data formats, and extensive free-text content. To address this, we developed a named entity recognition model leveraging retrieval-augmented generation (RAG) powered by generative artificial intelligence. The model identifies symptoms and triggers of agitation in dementia from nursing notes within residential aged care facilities (RACFs). By integrating RAG with few-shot learning, our re-ranking retrieval approach outperformed dense retrieval methods, achieving an accuracy of 0.87, an F1 score of 0.88, a recall of 0.90, and a precision of 0.86. This enhanced framework supports clinical decision-making, improving care quality and better management of dementia-related agitation in RACFs.

27 June 2024 in medRxiv
Verified
Preprint
Open access
PDF file
DVDinithi Vithanage
Dinithi Vithanage
CDChao Deng
Chao Deng
LWLei Wang
Lei Wang
MYMengyang Yin
Mengyang Yin
MAMohammad Alkhalaf
Mohammad Alkhalaf
Ping YuPing Yu
Ping Yu
Multi-label classification of unstructured electronic health records (EHR) is challenging due to the semantic complexity of textual data. Identifying the most effective machine learning method for EHR classification is useful in real-world clinical settings. Advances in natural language processing (NLP) using large language models (LLMs) offer promising solutions. Therefore, this experimental research aims to test the effects of zero-shot and few-shot learning prompting, with and without parameter-efficient fine-tuning (PEFT) and retrieval-augmented generation (RAG) of LLMs, on the multi-label classification of unstructured EHR data from residential aged care facilities (RACFs) in Australia. The four clinical tasks examined are agitation in dementia, depression in dementia, frailty index, and malnutrition risk factors, using the Llama 3.1-8B. Performance evaluation includes accuracy, macro-averaged precision, recall, and F1 score, supported by non-parametric statistical analyses. Results indicate that both zero-shot and few-shot learning, regardless of the use of PEFT and RAG, demonstrate equivalent performance across the clinical tasks when using the same prompting template. Few-shot learning consistently outperforms zero-shot learning when neither PEFT nor RAG is applied. Notably, PEFT significantly enhances model performance in both zero-shot and few-shot learning; however, RAG improves performance only in few-shot learning. After PEFT, the performance of zero-shot learning is equal to that of few-shot learning across clinical tasks. Additionally, few-shot learning with RAG surpasses zero-shot learning with RAG, while no significant difference exists between few-shot learning with RAG and zero-shot learning with PEFT. These findings offer crucial insights into LLMs for researchers, practitioners, and stakeholders utilizing LLMs in clinical document analysis.