Po-Chih Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Po-Chih Chen has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Jih-Jin Tsai
School of Medicine, College of Medicine, Kaohsiung Medical University
2 shared publications - Guey Chuen Perng
Institute of Basic Medical Sciences, College of Medicine, National Cheng Kung University, Tainan 701401, Taiwan
1 shared publication - Hui-Mien Hsiao
Department of Pediatrics, Children’s Healthcare of Atlanta, Atlanta, GA
1 shared publication - Hui-Ru Yang
Department of Laboratory Medicine, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan
1 shared publication - Kuei-Ying Lin
Department of Laboratory Medicine, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan
1 shared publication - Kulkanya Chokephaibulkit
Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand
1 shared publication
Latest Publications
- Open Access
- 22 April 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (9) , 1258
Comparative Evaluation of Five Multimodal Large Language Models for Medical Laboratory Image Recognition: Impact of Prompting Strategies on Diagnostic Accuracy
Background: Multimodal large language models (MLLMs) show promise in medical imaging, but their performance is highly dependent on prompt engineering. This study systematically evaluates how different prompting strategies affect diagnostic accuracy in clinical laboratory image interpretation. Methods: We evaluated five MLLMs (ChatGPT-4o, Gemini 2.0 Flash, Claude 3.5 Sonnet, Grok-2, and Perplexity Pro (Claude 3.5 Sonnet)) using 177 proficiency testing images across three domains: blood smears (n = 78), urinalysis (n = 50), and parasitology (n = 49). Three prompting approaches were compared: (1) complex multi-choice prompts with 20 diagnostic options, (2) zero-shot open-ended prompts, and (3) two-step descriptive-reasoning prompts. Images were sourced from the Taiwan Society of Laboratory Medicine external quality assurance archives with expert consensus diagnoses. Results: Zero-shot prompting significantly outperformed complex multi-choice prompts across all models and domains (p < 0.001). With zero-shot prompts, Gemini achieved 78.5% overall accuracy (urinalysis: 92.0%; parasitology: 75.5%; blood smears: 64.1%), representing a 17% improvement over complex prompts. Two-step descriptive-reasoning prompts further improved blood smear accuracy by 8–12% for top-performing models, but showed minimal benefit in urinalysis and parasitology. The re-query mechanism (“please reconsider”) improved urinalysis accuracy by 7.6% but had a negligible effect on blood smears and parasitology. Conclusions: Prompting strategy critically determines MLLM diagnostic performance. Zero-shot approaches with minimal constraints consistently outperform complex multi-choice formats. The remarkable performance of general-purpose models in structured domains like urinalysis (>90% accuracy) demonstrates the considerable progress of multimodal AI. However, complex morphological tasks like blood smear interpretation require either specialized prompting techniques or domain-specific fine-tuning. These findings provide evidence-based guidance for optimizing AI integration in clinical laboratories.
- 22 April 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (9) , 1258
Open AccessComparative Evaluation of Five Multimodal Large Language Models for Medical Laboratory Image Recognition: Impact of Prompting Strategies on Diagnostic Accuracy
Background: Multimodal large language models (MLLMs) show promise in medical imaging, but their performance is highly dependent on prompt engineering. This study systematically evaluates how different prompting strategies affect diagnostic accuracy in clinical laboratory image interpretation. Methods: We evaluated five MLLMs (ChatGPT-4o, Gemini 2.0 Flash, Claude 3.5 Sonnet, Grok-2, and Perplexity Pro (Claude 3.5 Sonnet)) using 177 proficiency testing images across three domains: blood smears (n = 78), urinalysis (n = 50), and parasitology (n = 49). Three prompting approaches were compared: (1) complex multi-choice prompts with 20 diagnostic options, (2) zero-shot open-ended prompts, and (3) two-step descriptive-reasoning prompts. Images were sourced from the Taiwan Society of Laboratory Medicine external quality assurance archives with expert consensus diagnoses. Results: Zero-shot prompting significantly outperformed complex multi-choice prompts across all models and domains (p < 0.001). With zero-shot prompts, Gemini achieved 78.5% overall accuracy (urinalysis: 92.0%; parasitology: 75.5%; blood smears: 64.1%), representing a 17% improvement over complex prompts. Two-step descriptive-reasoning prompts further improved blood smear accuracy by 8–12% for top-performing models, but showed minimal benefit in urinalysis and parasitology. The re-query mechanism (“please reconsider”) improved urinalysis accuracy by 7.6% but had a negligible effect on blood smears and parasitology. Conclusions: Prompting strategy critically determines MLLM diagnostic performance. Zero-shot approaches with minimal constraints consistently outperform complex multi-choice formats. The remarkable performance of general-purpose models in structured domains like urinalysis (>90% accuracy) demonstrates the considerable progress of multimodal AI. However, complex morphological tasks like blood smear interpretation require either specialized prompting techniques or domain-specific fine-tuning. These findings provide evidence-based guidance for optimizing AI integration in clinical laboratories.
- Open Access
- 5 December 2013
- Published by Springer Nature in Journal of Biomedical Science
- Vol. 20 (1) , 88
Role of cognitive parameters in dengue hemorrhagic fever and dengue shock syndrome
Dengue is becoming recognized as one of the most important vector-borne human diseases. It is predominant in tropical and subtropical zones but its geographical distribution is progressively expanding, making it an escalating global health problem of today. Dengue presents with spectrum of clinical manifestations, ranging from asymptomatic, undifferentiated mild fever, dengue fever (DF), to dengue hemorrhagic fever (DHF) with or without shock (DSS), a life-threatening illness characterized by plasma leakage due to increased vascular permeability. Currently, there are no antiviral modalities or vaccines available to treat and prevent dengue. Supportive care with close monitoring is the standard clinical practice. The mechanisms leading to DHF/DSS remains poorly understood. Multiple factors have been attributed to the pathological mechanism, but only a couple of these hypotheses are popular in scientific circles. The current discussion focuses on underappreciated factors, temperature, natural IgM, and endotoxin, which may be critical components playing roles in dengue pathogenesis.
- 5 December 2013
- Published by Springer Nature in Journal of Biomedical Science
- Vol. 20 (1) , 88
Open AccessRole of cognitive parameters in dengue hemorrhagic fever and dengue shock syndrome
Dengue is becoming recognized as one of the most important vector-borne human diseases. It is predominant in tropical and subtropical zones but its geographical distribution is progressively expanding, making it an escalating global health problem of today. Dengue presents with spectrum of clinical manifestations, ranging from asymptomatic, undifferentiated mild fever, dengue fever (DF), to dengue hemorrhagic fever (DHF) with or without shock (DSS), a life-threatening illness characterized by plasma leakage due to increased vascular permeability. Currently, there are no antiviral modalities or vaccines available to treat and prevent dengue. Supportive care with close monitoring is the standard clinical practice. The mechanisms leading to DHF/DSS remains poorly understood. Multiple factors have been attributed to the pathological mechanism, but only a couple of these hypotheses are popular in scientific circles. The current discussion focuses on underappreciated factors, temperature, natural IgM, and endotoxin, which may be critical components playing roles in dengue pathogenesis.