Thanawat Piyasawetkul

Mr. Thanawat Piyasawetkul

Graduate School in the Program of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences, Khon Kaen University, Khon Kaen, Thailand
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
29 July 2025 in AI
Verified
Journal Article
Open access
PDF file
Thanawat PiyasawetkulThanawat Piyasawetkul
Thanawat Piyasawetkul
STSuppachai Tiyaworanant
Suppachai Tiyaworanant
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
Trust in Traditional Thai Medicine (TTM) among Thai people has been reduced due to a lack of objective standards and the susceptibility of the general population to false information. The emergence of generative artificial intelligence (Gen AI) has significantly impacted various industries, including traditional medicine. However, previous Gen AI models have primarily focused on prescription generation based on Traditional Chinese Medicine (TCM), leaving TTM unexplored. To address this gap, we propose a novel fast-learning fine-tuned language model fortified with TTM knowledge. We utilized textual data from two TTM textbooks, Wat Ratcha-orasaram Ratchaworawihan (WRO), and Tamra Osot Phra Narai (NR), to fine-tune Unsloth’s Gemma-2 with 9 billion parameters. We developed two specialized TTM tasks: treatment prediction (TrP) and herbal recipe generation (HRG). The TrP and HRG models achieved precision, recall, and F1 scores of 26.54%, 28.14%, and 24.00%, and 32.51%, 24.42%, and 24.84%, respectively. Performance evaluation against TCM-based generative models showed comparable precision, recall, and F1 results with a smaller knowledge corpus. We further addressed the challenges of utilizing Thai, a low-resource and linguistically complex language. Unlike English or Chinese, Thai lacks explicit sentence boundary markers and employs an abugida writing system without spaces between words, complicating text segmentation and generation. These characteristics pose significant difficulties for machine understanding and limit model accuracy. Despite these obstacles, our work establishes a foundation for further development of AI-assisted TTM applications and highlights both the opportunities and challenges in applying language models to traditional medicine knowledge systems in Thai language contexts.

Verified
Journal Article
Open access
PDF file
Tarapong SrisongkramTarapong Srisongkram
Tarapong Srisongkram
HDHuynh Anh Duy
Huynh Anh Duy
SPSupreeya Paiboon
Supreeya Paiboon
Thanawat PiyasawetkulThanawat Piyasawetkul
Thanawat Piyasawetkul
DZDarlene Nabila Zetta
Darlene Nabila Zetta
Natthida WeerapreeyakulNatthida Weerapreeyakul
Natthida Weerapreeyakul
Mitogen-activated protein kinase 1 (MAPK1) plays a central role in cancer cell proliferation and survival. In this study, we developed an ensemble graph neural network framework integrating graph convolutional neural networks, graph message-passing neural networks, and attention-based graph neural networks to predict MAPK1 inhibitory activity using large-scale quantitative high-throughput screening data. Our models achieved recall ranging from 79.5% to 92.3%, areas under the precision–recall curve ranging from 80.4% to 90.3%, and Matthews correlation coefficients ranging from 0.523 to 0.725 on the independent test set, demonstrating a strong ability to identify MAPK1 inhibitors. The Y-randomization test further confirmed a true relationship between molecular graph structure and MAPK1 inhibitory activity. We also defined the model’s applicability domain to establish clear boundaries for confident prediction. To experimentally validate this approach, we evaluated the approved MAPK1 inhibitor ulixertinib and tested multiple FDA-approved kinase inhibitors, which underscored the model’s predictive ability in the drug development pipeline. Furthermore, we applied the developed models to predict and experimentally validate natural products as potential novel MAPK1 inhibitors. Finally, we developed a web server to deploy this model, enabling rapid and cost-effective identification of MAPK1 inhibitors for hit screening experiments prior to substantial in vitro testing.