Classification of Pancreatic Cancer and Normal Tissue in 2D and 3D Optical Coherence Tomography Images Using Convolutional Neural Networks: A Comparative Study.
Simple Summary: Surgeons treating pancreatic cancer need to remove all cancer tissue to give patients the best chance of recovery. This study looked at whether a special imaging method, called optical coherence tomography (OCT), combined with artificial intelligence (AI), could tell cancer tissue ap...
| Publicado en: | Cancers Vol. 18; no. 5; pp. 732 - 744 |
|---|---|
| Autores principales: | , , , , , , , , , , , , |
| Formato: | pictorial research tables/charts Journal Article |
| Publicado: |
MDPI
Mar2026
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=192641296&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 192641296 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 20726694 B74B jtl: Cancers issn: 20726694 maglogo: N pubinfo: dt: Mar2026 vid: 18 iid: 5 pid: 97109 pub: MDPI artinfo: ui: 192641296 192641296 192641296 10.3390/cancers18050732 192641296 ppf: 732 ppct: 12 formats: tig: atl: Classification of Pancreatic Cancer and Normal Tissue in 2D and 3D Optical Coherence Tomography Images Using Convolutional Neural Networks: A Comparative Study. aug: au: Druzenko, Maria Westerheide, Bastian Girmen, Caroline König, Niels Schmitt, Robert Warkentin, Svetlana Jöchle, Katharina Cammann, Sebastian Wiltberger, Georg von Websky, Martin W. Vogel, Thomas Vondran, Florian W. R. Amygdalos, Iakovos affil: Department of General, Visceral, Pediatric and Transplantation Surgery, University Hospital RWTH Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany sug: subj: Pancreatic Neoplasms Diagnosis Tomography, Optical Coherence Convolutional Neural Networks Utilization Imaging, Three-Dimensional Funding Source Germany Human Male Female Descriptive Statistics Prospective Studies Data Analysis Software Image Interpretation, Computer Assisted Sensitivity and Specificity Diagnosis, Computer Assisted Comparative Studies Predictive Value of Tests ROC Curve Biopsy Cancer Screening Machine Learning Artificial Intelligence Male Female ab: Simple Summary: Surgeons treating pancreatic cancer need to remove all cancer tissue to give patients the best chance of recovery. This study looked at whether a special imaging method, called optical coherence tomography (OCT), combined with artificial intelligence (AI), could tell cancer tissue apart from normal pancreatic tissue, which could be used to check if the entire tumor has been removed during surgery. Researchers scanned tissue that had already been removed from 27 patients with pancreatic cancer. They then trained computer programs to recognize differences between cancer and healthy tissue in these images. The best-performing program correctly identified cancer tissue most of the time and was also good at recognizing normal tissue. The results suggest that combining OCT with AI could one day help surgeons quickly check tissue during operations, possibly reducing the need for time-consuming laboratory tests. More research is needed to see how well this works during real surgeries on living patients. Background/Objectives: Early and complete (R0) surgical resection is essential for optimal outcomes in pancreatic cancer. Optical coherence tomography (OCT) combined with artificial intelligence (AI) may offer real-time intraoperative guidance, potentially reducing reliance on frozen sections. This ex vivo study evaluated convolutional neural networks (CNNs) for distinguishing pancreatic ductal adenocarcinoma (PDAC) from normal pancreatic tissue in OCT images obtained ex vivo. Methods: Between October 2020 and April 2021, OCT scans were obtained from resected pancreatic specimens of 27 adult patients. Tumor and adjacent normal tissue were imaged using a 1310 nm OCT system, followed by histopathological confirmation. A total of 25 PDAC and 30 non-malignant scans were preprocessed and analyzed using cross-validated CNN models (ResNet50, DenseNet121, and MobileNetV2) with both 2D and 3D inputs. Results: Using five-fold stratified cross-validation on 9040 2D and 3000 3D samples (224 px resolution), the 3D DenseNet121 model achieved the highest performance, with an F1-score of 0.74, sensitivity of 72%, and specificity of 81%. Other architectures demonstrated comparable results. Conclusions: AI-assisted OCT can accurately differentiate PDAC from normal pancreatic tissue ex vivo, supporting its potential as a rapid intraoperative diagnostic adjunct. Further studies are warranted to assess its in vivo performance and utility in evaluating resection margins. pubtype: Academic Journal doctype: pictorial research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|