Chia-Pei Tang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chia-Pei Tang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Tu-Liang Lin
Department of Management Information System, National Chiayi University, Chiayi 60054, Taiwan
4 shared publications - Hong-Yi Chang
Department of Management Information System, National Chiayi University, Chiayi City 60054, Taiwan
3 shared publications - Felix W. Leung
David Geffen School of Medicine, University of California Los Angeles, Los Angeles, USA
2 shared publications - Yu-Hsi Hsieh
Division of Gastroenterology, Department of Internal Medicine Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation Chiayi Taiwan
2 shared publications - Chen-Hung Hsieh
Department of Management Information System, National Chiayi University, Chiayi City 600023, Taiwan
1 shared publication - Cheng-Kui Huang
National Chung Cheng University, Taiwan
1 shared publication
Latest Publications
- Open Access
- 29 June 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (7) , 759
Cross-Domain Robust Pruning for Polyp Segmentation: Multi-Encoder Feature Fusion Beats Single-Encoder Baselines
Medical image segmentation requires dense pixel-level annotations, making large-scale dataset construction expensive and motivating research into data-efficient training. The main objective of this paper is to determine whether fusing two complementary pretrained image encoders into a single similarity space can make training-free dataset pruning robust across heterogeneous polyp segmentation domains and to quantify that robustness against a comprehensive panel of baselines. To achieve this, we propose Multi-Encoder Diverse Pruning (MEDP), a training-free dataset-pruning method. MEDP fuses features from an ImageNet-pretrained ResNet-18 and a self-supervised DINOv2 ViT-S/14 into a single 896-D similarity space. It partitions the training pool via Louvain modularity maximization and selects per-community samples via maximal-marginal-relevance (MMR) ranking, which effectively balances eigenvector centrality with feature-space diversity. We benchmarked MEDP against 12 baselines at a 20% retention ratio across three polyp segmentation settings (Kvasir-SEG, CVC-ClinicDB, and a Combined cross-domain pool) using a standard 5-level UNet. Based on approximately 330 controlled training runs, the results demonstrate that MEDP achieves the highest mean test Dice of 0.7324 on the most challenging Combined cross-domain pool, significantly outperforming uniform random sampling (Cohen’s d = +1.79, paired Wilcoxon p = 0.002). Conversely, all hand-crafted structure-aware variants failed to outperform uniform random sampling. These findings confirm that combining multi-encoder features with MMR diversity provides a simple and effective strategy for improving robustness across heterogeneous medical imaging settings and that the choice of pretrained image encoder is the dominant factor in segmentation-aware pruning.
- 29 June 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (7) , 759
Open AccessCross-Domain Robust Pruning for Polyp Segmentation: Multi-Encoder Feature Fusion Beats Single-Encoder Baselines
Medical image segmentation requires dense pixel-level annotations, making large-scale dataset construction expensive and motivating research into data-efficient training. The main objective of this paper is to determine whether fusing two complementary pretrained image encoders into a single similarity space can make training-free dataset pruning robust across heterogeneous polyp segmentation domains and to quantify that robustness against a comprehensive panel of baselines. To achieve this, we propose Multi-Encoder Diverse Pruning (MEDP), a training-free dataset-pruning method. MEDP fuses features from an ImageNet-pretrained ResNet-18 and a self-supervised DINOv2 ViT-S/14 into a single 896-D similarity space. It partitions the training pool via Louvain modularity maximization and selects per-community samples via maximal-marginal-relevance (MMR) ranking, which effectively balances eigenvector centrality with feature-space diversity. We benchmarked MEDP against 12 baselines at a 20% retention ratio across three polyp segmentation settings (Kvasir-SEG, CVC-ClinicDB, and a Combined cross-domain pool) using a standard 5-level UNet. Based on approximately 330 controlled training runs, the results demonstrate that MEDP achieves the highest mean test Dice of 0.7324 on the most challenging Combined cross-domain pool, significantly outperforming uniform random sampling (Cohen’s d = +1.79, paired Wilcoxon p = 0.002). Conversely, all hand-crafted structure-aware variants failed to outperform uniform random sampling. These findings confirm that combining multi-encoder features with MMR diversity provides a simple and effective strategy for improving robustness across heterogeneous medical imaging settings and that the choice of pretrained image encoder is the dominant factor in segmentation-aware pruning.
- Open Access
- 21 November 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (23) , 2949
Ordinal Regression Research Based on Dual Loss Function—An Example on Lumbar Vertebra Classification in CT Images
Background/Objectives: Some classification problems involve ordered categories (e.g., low–medium–high), which are better modeled as ordinal regression. This study aimed to propose and evaluate a dual loss framework—Ordinal Residual Dual Loss—for lumbar vertebra classification on CT images to assist L3 identification and sarcopenia detection. Methods: In this retrospective study, lumbar spine CT images were used to train a deep learning model based on a MobileNet-v3-Large network. The proposed framework combines standard cross-entropy loss for classification with an Ordinal Residual Loss defined on the difference between output probabilities and target ordinal probabilities. Results: Experimental results show that the Ordinal Residual Dual Loss approach outperforms using cross-entropy alone and also surpasses methods from previous studies in lumbar vertebra classification performance. Conclusions: Leveraging a dual loss design that incorporates ordinal information improves vertebral level classification on CT images and has potential to support more accurate automated L3 localization and sarcopenia assessment in clinical practice.
- 21 November 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (23) , 2949
Open AccessOrdinal Regression Research Based on Dual Loss Function—An Example on Lumbar Vertebra Classification in CT Images
Background/Objectives: Some classification problems involve ordered categories (e.g., low–medium–high), which are better modeled as ordinal regression. This study aimed to propose and evaluate a dual loss framework—Ordinal Residual Dual Loss—for lumbar vertebra classification on CT images to assist L3 identification and sarcopenia detection. Methods: In this retrospective study, lumbar spine CT images were used to train a deep learning model based on a MobileNet-v3-Large network. The proposed framework combines standard cross-entropy loss for classification with an Ordinal Residual Loss defined on the difference between output probabilities and target ordinal probabilities. Results: Experimental results show that the Ordinal Residual Dual Loss approach outperforms using cross-entropy alone and also surpasses methods from previous studies in lumbar vertebra classification performance. Conclusions: Leveraging a dual loss design that incorporates ordinal information improves vertebral level classification on CT images and has potential to support more accurate automated L3 localization and sarcopenia assessment in clinical practice.
- Open Access
- 4 May 2023
- Published by Wolters Kluwer Health in Clinical and Translational Gastroenterology
- Vol. 14 (7) , e00594
Randomized Trial Comparing Left Colon Mucus Production Using Water vs Saline During Water Exchange Colonoscopy
INTRODUCTION: Water-assisted colonoscopy increases left colon mucus production; however, the effect of saline on mucus production is unclear. We tested the hypothesis that saline infusion may reduce mucus production in a dose-related manner. METHODS: In a randomized trial, patients were assigned to colonoscopy with CO2 insufflation, water exchange (WE) with warm water, 25% saline, or 50% saline. The primary outcome was the Left Colon Mucus Scale (LCMS) score (5-point scale). Blood electrolytes were measured before and after saline infusion. RESULTS: A total of 296 patients with similar baseline demographics were included. The mean LCMS score for WE with water was significantly higher than that for WE with saline and CO2 (1.4 ± 0.8 [WE water] vs 0.7 ± 0.6 [WE 25% saline] vs 0.5 ± 0.5 [WE 50% saline] vs 0.2 ± 0.4 [CO2]; overall P < 0.0001), with no significant difference between the 25% and 50% saline groups. The left colon adenoma detection rate (ADR) was highest in the 50% saline group, followed by the 25% saline and the water groups (25.0% vs 18.7% vs 13.3%), but the difference was not significant. Logistic regression showed water infusion as the only predictor of moderate mucus production (odds ratio 33.3, 95% confidence interval 7.2–153.2). No acute electrolyte abnormalities were documented indicating a safe modification. DISCUSSION: The use of 25% and 50% saline significantly inhibited mucus production and numerically increased ADR in the left colon. Evaluation of the impact of mucus inhibition by saline on ADR may refine the outcomes of WE.
- 4 May 2023
- Published by Wolters Kluwer Health in Clinical and Translational Gastroenterology
- Vol. 14 (7) , e00594
Open AccessRandomized Trial Comparing Left Colon Mucus Production Using Water vs Saline During Water Exchange Colonoscopy
INTRODUCTION: Water-assisted colonoscopy increases left colon mucus production; however, the effect of saline on mucus production is unclear. We tested the hypothesis that saline infusion may reduce mucus production in a dose-related manner. METHODS: In a randomized trial, patients were assigned to colonoscopy with CO2 insufflation, water exchange (WE) with warm water, 25% saline, or 50% saline. The primary outcome was the Left Colon Mucus Scale (LCMS) score (5-point scale). Blood electrolytes were measured before and after saline infusion. RESULTS: A total of 296 patients with similar baseline demographics were included. The mean LCMS score for WE with water was significantly higher than that for WE with saline and CO2 (1.4 ± 0.8 [WE water] vs 0.7 ± 0.6 [WE 25% saline] vs 0.5 ± 0.5 [WE 50% saline] vs 0.2 ± 0.4 [CO2]; overall P < 0.0001), with no significant difference between the 25% and 50% saline groups. The left colon adenoma detection rate (ADR) was highest in the 50% saline group, followed by the 25% saline and the water groups (25.0% vs 18.7% vs 13.3%), but the difference was not significant. Logistic regression showed water infusion as the only predictor of moderate mucus production (odds ratio 33.3, 95% confidence interval 7.2–153.2). No acute electrolyte abnormalities were documented indicating a safe modification. DISCUSSION: The use of 25% and 50% saline significantly inhibited mucus production and numerically increased ADR in the left colon. Evaluation of the impact of mucus inhibition by saline on ADR may refine the outcomes of WE.
- Open Access
- 4 January 2023
- Published by MDPI AG in Diagnostics
- Vol. 13 (2) , 170
A Novel Computer-Aided Detection/Diagnosis System for Detection and Classification of Polyps in Colonoscopy
Using a deep learning algorithm in the development of a computer-aided system for colon polyp detection is effective in reducing the miss rate. This study aimed to develop a system for colon polyp detection and classification. We used a data augmentation technique and conditional GAN to generate polyp images for YOLO training to improve the polyp detection ability. After testing the model five times, a model with 300 GANs (GAN 300) achieved the highest average precision (AP) of 54.60% for SSA and 75.41% for TA. These results were better than those of the data augmentation method, which showed AP of 53.56% for SSA and 72.55% for TA. The AP, mAP, and IoU for the 300 GAN model for the HP were 80.97%, 70.07%, and 57.24%, and the data increased in comparison with the data augmentation technique by 76.98%, 67.70%, and 55.26%, respectively. We also used Gaussian blurring to simulate the blurred images during colonoscopy and then applied DeblurGAN-v2 to deblur the images. Further, we trained the dataset using YOLO to classify polyps. After using DeblurGAN-v2, the mAP increased from 25.64% to 30.74%. This method effectively improved the accuracy of polyp detection and classification.
- 4 January 2023
- Published by MDPI AG in Diagnostics
- Vol. 13 (2) , 170
Open AccessA Novel Computer-Aided Detection/Diagnosis System for Detection and Classification of Polyps in Colonoscopy
Using a deep learning algorithm in the development of a computer-aided system for colon polyp detection is effective in reducing the miss rate. This study aimed to develop a system for colon polyp detection and classification. We used a data augmentation technique and conditional GAN to generate polyp images for YOLO training to improve the polyp detection ability. After testing the model five times, a model with 300 GANs (GAN 300) achieved the highest average precision (AP) of 54.60% for SSA and 75.41% for TA. These results were better than those of the data augmentation method, which showed AP of 53.56% for SSA and 72.55% for TA. The AP, mAP, and IoU for the 300 GAN model for the HP were 80.97%, 70.07%, and 57.24%, and the data increased in comparison with the data augmentation technique by 76.98%, 67.70%, and 55.26%, respectively. We also used Gaussian blurring to simulate the blurred images during colonoscopy and then applied DeblurGAN-v2 to deblur the images. Further, we trained the dataset using YOLO to classify polyps. After using DeblurGAN-v2, the mAP increased from 25.64% to 30.74%. This method effectively improved the accuracy of polyp detection and classification.
- Open Access
- 12 April 2022
- Published by MDPI AG in Diagnostics
- Vol. 12 (4) , 968
Computer-Aided Image Enhanced Endoscopy Automated System to Boost Polyp and Adenoma Detection Accuracy
Colonoscopy is the gold standard to detect colon polyps prematurely. Early detection, characterization and resection of polyps decrease colon cancer incidence. Colon polyp missing rate remains high despite novel methods development. Narrowed-band imaging (NBI) is one of the image enhance techniques used to boost polyp detection and characterization, which uses special filters to enhance the contrast of the mucosa surface and vascular pattern of the polyp. However, the single-button-activated system is not convenient for a full-time colonoscopy operation. We selected three methods to simulate the NBI system: Color Transfer with Mean Shift (CTMS), Multi-scale Retinex with Color Restoration (MSRCR), and Gamma and Sigmoid Conversions (GSC). The results show that the classification accuracy using the original images is the lowest. All color transfer methods outperform the original images approach. Our results verified that the color transfer has a positive impact on the polyp identification and classification task. Combined analysis results of the mAP and the accuracy show an excellent performance of the MSRCR method.
- 12 April 2022
- Published by MDPI AG in Diagnostics
- Vol. 12 (4) , 968
Open AccessComputer-Aided Image Enhanced Endoscopy Automated System to Boost Polyp and Adenoma Detection Accuracy
Colonoscopy is the gold standard to detect colon polyps prematurely. Early detection, characterization and resection of polyps decrease colon cancer incidence. Colon polyp missing rate remains high despite novel methods development. Narrowed-band imaging (NBI) is one of the image enhance techniques used to boost polyp detection and characterization, which uses special filters to enhance the contrast of the mucosa surface and vascular pattern of the polyp. However, the single-button-activated system is not convenient for a full-time colonoscopy operation. We selected three methods to simulate the NBI system: Color Transfer with Mean Shift (CTMS), Multi-scale Retinex with Color Restoration (MSRCR), and Gamma and Sigmoid Conversions (GSC). The results show that the classification accuracy using the original images is the lowest. All color transfer methods outperform the original images approach. Our results verified that the color transfer has a positive impact on the polyp identification and classification task. Combined analysis results of the mAP and the accuracy show an excellent performance of the MSRCR method.
- Open Access
Computer-Aided Colon Polyp Detection on High Resolution Colonoscopy Using Transfer Learning Techniques
Colonoscopies reduce the incidence of colorectal cancer through early recognition and resecting of the colon polyps. However, the colon polyp miss detection rate is as high as 26% in conventional colonoscopy. The search for methods to decrease the polyp miss rate is nowadays a paramount task. A number of algorithms or systems have been developed to enhance polyp detection, but few are suitable for real-time detection or classification due to their limited computational ability. Recent studies indicate that the automated colon polyp detection system is developing at an astonishing speed. Real-time detection with classification is still a yet to be explored field. Newer image pattern recognition algorithms with convolutional neuro-network (CNN) transfer learning has shed light on this topic. We proposed a study using real-time colonoscopies with the CNN transfer learning approach. Several multi-class classifiers were trained and mAP ranged from 38% to 49%. Based on an Inception v2 model, a detector adopting a Faster R-CNN was trained. The mAP of the detector was 77%, which was an improvement of 35% compared to the same type of multi-class classifier. Therefore, our results indicated that the polyp detection model could attain a high accuracy, but the polyp type classification still leaves room for improvement.
Open AccessComputer-Aided Colon Polyp Detection on High Resolution Colonoscopy Using Transfer Learning Techniques
Colonoscopies reduce the incidence of colorectal cancer through early recognition and resecting of the colon polyps. However, the colon polyp miss detection rate is as high as 26% in conventional colonoscopy. The search for methods to decrease the polyp miss rate is nowadays a paramount task. A number of algorithms or systems have been developed to enhance polyp detection, but few are suitable for real-time detection or classification due to their limited computational ability. Recent studies indicate that the automated colon polyp detection system is developing at an astonishing speed. Real-time detection with classification is still a yet to be explored field. Newer image pattern recognition algorithms with convolutional neuro-network (CNN) transfer learning has shed light on this topic. We proposed a study using real-time colonoscopies with the CNN transfer learning approach. Several multi-class classifiers were trained and mAP ranged from 38% to 49%. Based on an Inception v2 model, a detector adopting a Faster R-CNN was trained. The mAP of the detector was 77%, which was an improvement of 35% compared to the same type of multi-class classifier. Therefore, our results indicated that the polyp detection model could attain a high accuracy, but the polyp type classification still leaves room for improvement.
- Open Access
- 18 June 2021
- Published by MDPI AG in Diagnostics
- Vol. 11 (6) , 1113
Computer-Aided Detection False Positives in Colonoscopy
Randomized control trials and meta-analyses comparing colonoscopies with and without computer-aided detection (CADe) assistance showed significant increases in adenoma detection rates (ADRs) with CADe. A major limitation of CADe is its false positives (FPs), ranked 3rd in importance among 59 research questions in a modified Delphi consensus review. The definition of FPs varies. One commonly used definition defines an FP as an activation of the CADe system, irrespective of the number of frames or duration of time, not due to any polypoid or nonpolypoid lesions. Although only 0.07 to 0.2 FPs were observed per colonoscopy, video analysis studies using FPs as the primary outcome showed much higher numbers of 26 to 27 per colonoscopy. Most FPs were of short duration (91% < 0.5 s). A higher number of FPs was also associated with suboptimal bowel preparation. The appearance of FPs can lead to user fatigue. The polypectomy of FPs results in increased procedure time and added use of resources. Re-training the CADe algorithms is one way to reduce FPs but is not practical in the clinical setting during colonoscopy. Water exchange (WE) is an emerging method that the colonoscopist can use to provide salvage cleaning during insertion. We discuss the potential of WE for reducing FPs as well as the augmentation of ADRs through CADe.
- 18 June 2021
- Published by MDPI AG in Diagnostics
- Vol. 11 (6) , 1113
Open AccessComputer-Aided Detection False Positives in Colonoscopy
Randomized control trials and meta-analyses comparing colonoscopies with and without computer-aided detection (CADe) assistance showed significant increases in adenoma detection rates (ADRs) with CADe. A major limitation of CADe is its false positives (FPs), ranked 3rd in importance among 59 research questions in a modified Delphi consensus review. The definition of FPs varies. One commonly used definition defines an FP as an activation of the CADe system, irrespective of the number of frames or duration of time, not due to any polypoid or nonpolypoid lesions. Although only 0.07 to 0.2 FPs were observed per colonoscopy, video analysis studies using FPs as the primary outcome showed much higher numbers of 26 to 27 per colonoscopy. Most FPs were of short duration (91% < 0.5 s). A higher number of FPs was also associated with suboptimal bowel preparation. The appearance of FPs can lead to user fatigue. The polypectomy of FPs results in increased procedure time and added use of resources. Re-training the CADe algorithms is one way to reduce FPs but is not practical in the clinical setting during colonoscopy. Water exchange (WE) is an emerging method that the colonoscopist can use to provide salvage cleaning during insertion. We discuss the potential of WE for reducing FPs as well as the augmentation of ADRs through CADe.
- 1 December 2019
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 100, 103302
A trade-off dual-factor model to investigate discontinuous intention of health app users: From the perspective of information disclosure
- 1 December 2019
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 100, 103302
A trade-off dual-factor model to investigate discontinuous intention of health app users: From the perspective of information disclosure