EUVIMEDEUVIMED entdecken
Crossref

Deep learning via pathology-image integration for margin prediction after pancreatoduodenectomy

Ansley B Ricker, Sagar Satyanarayana, Trenton Pritt, Stephanie Jensen, Katheryn Peterson, Rohit Mantha, Allyson Cochran, John B Martinie, Erin H Baker

HPB · 2026-07 · 28 (7) · 889-896

DOI: 10.1016/j.hpb.2026.04.445 ↗

Kurz gefasst

Fachartikel in der Zeitschrift HPB, 2026-07. Verfasst von Ansley B Ricker und 8 weiteren Autor:innen. Beteiligte Einrichtung: Division of HPB Surgery, Department of Surgery, Atrium Health Carolinas Medical Center, Charlotte, NC, USA. Hauptthemen laut MEDLINE: Pancreaticoduodenectomy, Deep Learning, Margins of Excision, Carcinoma, Pancreatic Ductal. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 22 Literaturangaben. Gefördert von: Atrium Health, Wake Forest University School of Medicine.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: R1 margin status after pancreatoduodenectomy (PD) for pancreatic ductal adenocarcinoma (PDAC) remains common and is associated with inferior survival. Preoperative imaging inadequately predicts microscopic margin involvement, particularly at the superior mesenteric artery interface. This study evaluated whether deep learning applied to preoperative CT imaging could predict pathological margin status. METHODS: Patients who underwent PD for PDAC between 2007-2021 were identified from a prospectively maintained database. Preoperative triphasic CT scans were processed into standardized axial slices and labeled according to pathological margin status. Two convolutional neural network models were developed using an attention-based multiple instance learning approach. Model 1 used all slices from each CT scan. Model 2 incorporated noise-reduction techniques, including selective slice extraction, grayscale normalization, artifact removal, and cropping to the pancreas and surrounding tissue. RESULTS: One hundred patients met inclusion criteria (50 R0, 50 R1/R2), yielding 14,445 CT slices. Model 1 achieved an AUC of 0.69. After noise reduction, Model 2 demonstrated improved performance with an AUC of 0.78 and 88.2% sensitivity for identifying margin-positive cases. CONCLUSION: Deep learning applied to preoperative CT imaging shows feasibility for predicting pathological margin status after PD. Noise-reduction preprocessing improved performance, supporting imaging in preoperative risk stratification and surgical planning.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PancreaticoduodenectomyDeep LearningMargins of ExcisionCarcinoma, Pancreatic DuctalPancreatic NeoplasmsTomography, X-Ray ComputedHumansFemaleMaleAgedMiddle AgedConvolutional Neural NetworksPredictive Value of TestsMultiple-Instance Learning AlgorithmsPredictive Learning Models

Autor:innen und Institutionen

  • Division of HPB Surgery, Department of Surgery, Atrium Health Carolinas Medical Center, Charlotte, NC, USA.
  • Center for Surgical Outcomes Science, Atrium Health, Charlotte, NC, USA.
  • Division of HPB Surgery, Department of Surgery, Atrium Health Carolinas Medical Center, Charlotte, NC, USA. Electronic address: Erin.Baker@advocatehealth.org.

Förderung

  • Atrium Health
  • Wake Forest University School of Medicine

Zitieren

APA

Ricker, A. B., Satyanarayana, S., Pritt, T., Jensen, S., Peterson, K., Mantha, R., Cochran, A., Martinie, J. B., & Baker, E. H. (2026). Deep learning via pathology-image integration for margin prediction after pancreatoduodenectomy. HPB, 28(7), 889-896. https://doi.org/10.1016/j.hpb.2026.04.445

Vancouver

Ricker AB, Satyanarayana S, Pritt T, Jensen S, Peterson K, Mantha R, et al. Deep learning via pathology-image integration for margin prediction after pancreatoduodenectomy. HPB. 2026;28(7):889-896. doi:10.1016/j.hpb.2026.04.445

BibTeX
@article{Ricker2026,
  author = {Ricker, Ansley B and Satyanarayana, Sagar and Pritt, Trenton and Jensen, Stephanie and Peterson, Katheryn and Mantha, Rohit and Cochran, Allyson and Martinie, John B and Baker, Erin H},
  title = {Deep learning via pathology-image integration for margin prediction after pancreatoduodenectomy},
  journal = {HPB},
  year = {2026},
  volume = {28},
  number = {7},
  pages = {889--896},
  doi = {10.1016/j.hpb.2026.04.445}
}
Kostenlos starten →