A Visual Aid Tool for Detection of Pancreatic Tumour-Vessel Contact on Staging CT: A Retrospective Cohort Study

被引:1
作者
Policelli, Robert [1 ]
Dammak, Salma [2 ]
Ward, Aaron D. [1 ,2 ,3 ]
Kassam, Zahra [4 ,5 ]
Johnson, Carol [6 ]
Ramsewak, Darryl [4 ,6 ]
Syed, Zafir [4 ]
Siddiqi, Lubna [4 ]
Siddique, Naman [4 ]
Kim, Dongkeun [4 ]
Marshall, Harry [4 ,5 ,6 ]
机构
[1] Western Univ, Dept Med Biophys, 1151 Richmond St, London, ON N6A 3K7, Canada
[2] Western Univ, Sch Biomed Engn, London, ON, Canada
[3] Western Univ, Dept Oncol, London, ON, Canada
[4] Western Univ, Dept Med Imaging, London, ON, Canada
[5] St Josephs Hlth Care London, London, ON, Canada
[6] London Hlth Sci Ctr, London, ON, Canada
来源
CANADIAN ASSOCIATION OF RADIOLOGISTS JOURNAL-JOURNAL DE L ASSOCIATION CANADIENNE DES RADIOLOGISTES | 2024年 / 75卷 / 03期
关键词
pancreatic cancer; CT; radiology residents; visual aid tool; superior mesenteric artery; DUCTAL ADENOCARCINOMA; CANCER;
D O I
10.1177/08465371231217155
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Purpose: In pancreatic adenocarcinoma, the difficult distinction between normal and affected pancreas on CT studies may lead to discordance between the pre-surgical assessment of vessel involvement and intraoperative findings. We hypothesize that a visual aid tool could improve the performance of radiology residents when detecting vascular invasion in pancreatic adenocarcinoma patients. Methods: This study consisted of 94 pancreatic adenocarcinoma patient CTs. The visual aid compared the estimated body fat density of each patient with the densities surrounding the superior mesenteric artery and mapped them onto the CT scan. Four radiology residents annotated the locations of perceived vascular invasion on each scan with the visual aid overlaid on alternating scans. Using 3 expert radiologists as the reference standard, we quantified the area under the receiver operating characteristic curve to determine the performance of the tool. We then used sensitivity, specificity, balanced accuracy ((sensitivity + specificity)/2), and spatial metrics to determine the performance of the residents with and without the tool. Results: The mean area under the curve was 0.80. Radiology residents' sensitivity/specificity/balanced accuracy for predicting vascular invasion were 50%/85%/68% without the tool and 81%/79%/80% with it compared to expert radiologists, and 58%/85%/72% without the tool and 78%/77%/77% with it compared to the surgical pathology. The tool was not found to impact the spatial metrics calculated on the resident annotations of vascular invasion. Conclusion: The improvements provided by the visual aid were predominantly reflected by increased sensitivity and accuracy, indicating the potential of this tool as a learning aid for trainees. Visual Abstract This is a visual representation of the abstract. Objectif : Il est difficile de distinguer un pancr & eacute;as sain d'un pancr & eacute;as atteint sur des r & eacute;sultats de tomodensitom & eacute;trie (TDM), ce qui peut mener & agrave; des incoh & eacute;rences entre les r & eacute;sultats de l'& eacute;valuation pr & eacute;op & eacute;ratoire et les observations lors de l'intervention chirurgicale en ce qui concerne l'envahissement des vaisseaux adjacents dans les cas d'ad & eacute;nocarcinomes pancr & eacute;atiques. Nous avons & eacute;mis l'hypoth & egrave;se qu'un outil de visualisation pourrait am & eacute;liorer les performances des r & eacute;sidents en radiologie au moment de d & eacute;terminer la pr & eacute;sence d'un envahissement vasculaire chez les patients atteints d'un ad & eacute;nocarcinome pancr & eacute;atique. M & eacute;thodes : Dans le cadre de cette & eacute;tude, 94 images de TDM de patients atteints d'un ad & eacute;nocarcinome pancr & eacute;atique ont & eacute;t & eacute; analys & eacute;es. L'outil de visualisation a & eacute;tabli une comparaison entre la valeur estim & eacute;e de la densit & eacute; de graisse corporelle de chaque patient et la densit & eacute; de graisse autour de l'art & egrave;re m & eacute;sent & eacute;rique sup & eacute;rieure, et a permis de repr & eacute;senter ces donn & eacute;es sur l'image de TDM. Quatre r & eacute;sidents en radiologie ont annot & eacute; les emplacements de l'envahissement vasculaire per & ccedil;u sur chaque image; la moiti & eacute; des images & eacute;taient modifi & eacute;es au moyen de l'outil de visualisation. Afin d'obtenir une norme de r & eacute;f & eacute;rence, nous avons fait appel & agrave; trois radiologistes experts, et & agrave; l'aide de leurs annotations, l'aire sous la courbe de fonction d'efficacit & eacute; du r & eacute;cepteur a & eacute;t & eacute; calcul & eacute;e dans le but de d & eacute;terminer la performance de l'outil. Ensuite, des m & eacute;triques de sensibilit & eacute; diagnostique, de sp & eacute;cificit & eacute;, de pr & eacute;cision pond & eacute;r & eacute;e ((sensibilit & eacute;+sp & eacute;cificit & eacute;)/2) et d'analyse spatiale ont & eacute;t & eacute; calcul & eacute;es afin de d & eacute;terminer les performances des r & eacute;sidents lors de l'& eacute;valuation d'images modifi & eacute;es par l'outil ou non. R & eacute;sultats : L'aire sous la courbe moyenne & eacute;tait de 0,80. La sensibilit & eacute;, sp & eacute;cificit & eacute; et pr & eacute;cision pond & eacute;r & eacute;e des r & eacute;sidents en radiologie lors de la d & eacute;termination de la pr & eacute;sence d'un envahissement vasculaire en comparaison avec les annotations des radiologistes experts & eacute;tait de 50/85/68 % pour les images non modifi & eacute;es par l'outil et de 81/79/80 % pour les images modifi & eacute;es; leurs r & eacute;sultats en comparaison avec les r & eacute;sultats observ & eacute;s lors de l'intervention chirurgicale ont & eacute;t & eacute; de 58/85/72 % pour les images non modifi & eacute;es par l'outil et de 78/77/77 % pour les images modifi & eacute;es. Le recours & agrave; l'outil n'a pas eu d'effet sur les m & eacute;triques d'analyse spatiale calcul & eacute;es & agrave; partir des annotations des r & eacute;sidents au sujet de l'envahissement vasculaire. Conclusion : Les avantages que procure l'outil de visualisation se traduisaient principalement par une sensibilit & eacute; et
引用
收藏
页码:575 / 583
页数:9
相关论文
共 22 条
  • [1] Pancreatic Ductal Adenocarcinoma Radiology Reporting Template: Consensus Statement of the Society of Abdominal Radiology and the American Pancreatic Association
    Al-Hawary, Mahmoud M.
    Francis, Isaac R.
    Chari, Suresh T.
    Fishman, Elliot K.
    Hough, David M.
    Lu, David S.
    Macari, Michael
    Megibow, Alec J.
    Miller, Frank H.
    Mortele, Koenraad J.
    Merchant, Nipun B.
    Minter, Rebecca M.
    Tamm, Eric P.
    Sahani, Dushyant V.
    Simeone, Diane M.
    [J]. RADIOLOGY, 2014, 270 (01) : 248 - 260
  • [2] Bachmann Jeannine, 2006, HPB (Oxford), V8, P346, DOI 10.1080/13651820600803981
  • [3] Ultrasonography, computed tomography and magnetic resonance imaging for diagnosis and determining resectability of pancreatic adenocarcinoma -: A meta-analysis
    Bipat, S
    Phoa, SSKS
    van Delden, OM
    Bossuyt, PMM
    Gouma, DJ
    Laméris, JS
    Stoker, J
    [J]. JOURNAL OF COMPUTER ASSISTED TOMOGRAPHY, 2005, 29 (04) : 438 - 445
  • [4] Projected estimates of cancer in Canada in 2020
    Brenner, Darren R.
    Weir, Hannah K.
    Demers, Alain A.
    Ellison, Larry F.
    Louzado, Cheryl
    Shaw, Amanda
    Turner, Donna
    Woods, Ryan R.
    Smith, Leah M.
    [J]. CANADIAN MEDICAL ASSOCIATION JOURNAL, 2020, 192 (09) : E199 - E205
  • [5] Busireddy Kiran K, 2014, World J Gastrointest Pathophysiol, V5, P252, DOI 10.4291/wjgp.v5.i3.252
  • [6] Pretreatment Assessment of Resectable and Borderline Resectable Pancreatic Cancer: Expert Consensus Statement
    Callery, Mark P.
    Chang, Kenneth J.
    Fishman, Elliot K.
    Talamonti, Mark S.
    Traverso, L. William
    Linehan, David C.
    [J]. ANNALS OF SURGICAL ONCOLOGY, 2009, 16 (07) : 1727 - 1733
  • [7] Visualization of the fat planes between the pancreas and the adjacent organs and blood vessels using multi-detector computed tomography
    Djuric-Stefanovic, A.
    Gordanic, N.
    Saponjski, D.
    Koljensic, K.
    Djokic-Kovac, J.
    Knezevic, S.
    [J]. SURGICAL AND RADIOLOGIC ANATOMY, 2019, 41 (07) : 745 - 753
  • [8] 3D Slicer as an image computing platform for the Quantitative Imaging Network
    Fedorov, Andriy
    Beichel, Reinhard
    Kalpathy-Cramer, Jayashree
    Finet, Julien
    Fillion-Robin, Jean-Christophe
    Pujol, Sonia
    Bauer, Christian
    Jennings, Dominique
    Fennessy, Fiona
    Sonka, Milan
    Buatti, John
    Aylward, Stephen
    Miller, James V.
    Pieper, Steve
    Kikinis, Ron
    [J]. MAGNETIC RESONANCE IMAGING, 2012, 30 (09) : 1323 - 1341
  • [9] Pancreatic cancer: A review of epidemiology, trend, and risk factors
    Hu, Jian-Xiong
    Zhao, Cheng-Fei
    Chen, Wen-Biao
    Liu, Qi-Cai
    Li, Qu-Wen
    Lin, Yan-Ya
    Gao, Feng
    [J]. WORLD JOURNAL OF GASTROENTEROLOGY, 2021, 27 (27) : 4298 - 4321
  • [10] Preoperative CT Classification of the Resectability of Pancreatic Cancer: Interobserver Agreement
    Joo, Ijin
    Lee, Jeong Min
    Lee, Eun Sun
    Son, Jee-Young
    Lee, Dong Ho
    Ahn, Su Joa
    Chang, Won
    Lee, Sang Min
    Kang, Hyo-Jin
    Yang, Hyun Kyung
    [J]. RADIOLOGY, 2019, 293 (02) : 343 - 349