>
Publications

Every line of evidence, published internationally.

Five papers spanning algorithm development, nationwide population-based validation and real-world application — click "Read the paper" to open the full article on PubMed.

PubMed

5 international publications

Representative publications related to PANCREASaver®, all authored by the NTU team (Kao-Lang Liu, Po-Ting Chen, Wei-Chih Liao, Weichung Wang). Chinese summaries are translated from the original abstracts; English summaries are condensed from the originals.

The Lancet Digital Health · 2020 Open Access

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

92.1%Sensitivity for tumors <2cm 11/12Radiologist-missed cancers recovered AUC 0.997Local test sets

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation.

Approximately 40% of pancreatic tumors smaller than 2 cm evade CT detection. A CNN distinguished pancreatic cancer tissue with 97.3–99.0% sensitivity and AUC 0.997–0.999 on local test sets, and 79.0% sensitivity / AUC 0.920 on a US dataset. CNN outperformed radiologists (98.3% vs 92.9%), correctly classified 11 of 12 radiologist-missed cancers (92%), and reached 92.1% sensitivity for tumors <2 cm.

DOI: 10.1016/S2589-7500(20)30078-9 · PMID 33328124
Read the paper ↗
Radiology · 2023 Nationwide Study

Pancreatic Cancer Detection on CT Scans with Deep Learning: A Nationwide Population-based Study

Pancreatic Cancer Detection on CT Scans with Deep Learning: A Nationwide Population-based Study

1,473Nationwide real-world CT AUC 0.95Nationwide validation 89.7%Sensitivity

Pancreatic Cancer Detection on CT Scans with Deep Learning: A Nationwide Population-based Study.

~40% of pancreatic tumors <2 cm are missed at abdominal CT. An end-to-end DL tool (segmentation CNN + ensemble of five CNNs) achieved 89.9% sensitivity / 95.9% specificity (AUC 0.96) internally, and across 1,473 nationwide real-world CT studies (669 malignant, 804 control): 89.7% sensitivity, 92.8% specificity, AUC 0.95, with 74.7% sensitivity for tumors <2 cm.

DOI: 10.1148/radiol.220152 · PMID 36098642
Read the paper ↗
Radiology: Imaging Cancer · 2021 Radiomics

Radiomic Features at CT Can Distinguish Pancreatic Cancer from Noncancerous Pancreas

Radiomic Features at CT Can Distinguish Pancreatic Cancer from Noncancerous Pancreas

Taiwan / USTwo-population validation AUC 0.98Taiwan test set 94.7%Sensitivity · Taiwan

Radiomic Features at CT Can Distinguish Pancreatic Cancer from Noncancerous Pancreas.

XGBoost radiomic analysis of CT patches distinguished PDAC from noncancerous pancreas. The generalized model (trained on Taiwanese + U.S. data) reached 94.7% sensitivity / 95.4% specificity / AUC 0.98 on the Taiwanese test set, and 80.6% / 100% / AUC 0.91 on the U.S. test set. PDACs showed lower intensity and higher heterogeneity radiomic features.

DOI: 10.1148/rycan.2021210010 · PMID 34241550
Read the paper ↗
BMC Cancer · 2023 Real-world Dataset

Detection of pancreatic cancer with two- and three-dimensional radiomic analysis in a nationwide population-based real-world dataset

Detection of pancreatic cancer with two- and three-dimensional radiomic analysis in a nationwide population-based real-world dataset

1,477Nationwide real-world CT AUC 0.947Validation result 91.8%Sensitivity

Detection of pancreatic cancer with two- and three-dimensional radiomic analysis in a nationwide population-based real-world dataset.

An automatic end-to-end CAD tool combining 2D and 3D radiomic machine-learning analysis reached 91.8% sensitivity / 82.2% specificity / AUC 0.947 in 1,477 nationwide CT studies (671 PC, 806 controls), with 70.7% sensitivity for tumors <2 cm. Running 2D and 3D analyses in series raised specificity to 95.2%.

DOI: 10.1186/s12885-023-10536-8 · PMID 36650440
Read the paper ↗
J Gastroenterol Hepatol · 2021 Review

Applications of artificial intelligence in pancreatic and biliary diseases

Applications of artificial intelligence in pancreatic and biliary diseases

ReviewReview ML＋DLMethodology overview HepatobiliaryApplication area

Applications of artificial intelligence in pancreatic and biliary diseases.

A concise review of major AI methodologies (machine learning, deep learning) and the current landscape of AI research in pancreatobiliary diseases — where diagnosis and treatment selection are often complex — covering detection/diagnosis, risk stratification and prognosis prediction to supplement clinicians.

DOI: 10.1111/jgh.15380 · PMID 33624891
Read the paper ↗

Chinese summaries are translated from the original abstracts (English summaries condensed); the PubMed originals are authoritative.

Back to Clinical Evidence