1
|
Xin H, Zhang Y, Lai Q, Liao N, Zhang J, Liu Y, Chen Z, He P, He J, Liu J, Zhou Y, Yang W, Zhou Y. Automatic origin prediction of liver metastases via hierarchical artificial-intelligence system trained on multiphasic CT data: a retrospective, multicentre study. EClinicalMedicine 2024; 69:102464. [PMID: 38333364 PMCID: PMC10847157 DOI: 10.1016/j.eclinm.2024.102464] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/12/2023] [Revised: 12/22/2023] [Accepted: 01/17/2024] [Indexed: 02/10/2024] Open
Abstract
Background Currently, the diagnostic testing for the primary origin of liver metastases (LMs) can be laborious, complicating clinical decision-making. Directly classifying the primary origin of LMs at computed tomography (CT) images has proven to be challenging, despite its potential to streamline the entire diagnostic workflow. Methods We developed ALMSS, an artificial intelligence (AI)-based LMs screening system, to provide automated liver contrast-enhanced CT analysis for distinguishing LMs from hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC), as well as subtyping primary origin of LMs as six organ systems. We processed a CECT dataset between January 1, 2013 and June 30, 2022 (n = 3105: 840 HCC, 354 ICC, and 1911 LMs) for training and internally testing ALMSS, and two additional cohorts (n = 622) for external validation of its diagnostic performance. The performance of radiologists with and without the assistance of ALMSS in diagnosing and subtyping LMs was assessed. Findings ALMSS achieved average area under the curve (AUC) of 0.917 (95% confidence interval [CI]: 0.899-0.931) and 0.923 (95% [CI]: 0.905-0.937) for differentiating LMs, HCC and ICC on both the internal testing set and external testing set, outperformed that of two radiologists. Moreover, ALMSS yielded average AUC of 0.815 (95% [CI]: 0.794-0.836) and 0.818 (95% [CI]: 0.790-0.842) for predicting six primary origins on both two testing sets. Interestingly, ALMSS assigned origin diagnoses for LMs with pathological phenotypes beyond the training categories with average AUC of 0.761 (95% [CI]: 0.657-0.842), which verify the model's diagnostic expandability. Interpretation Our study established an AI-based diagnostic system that effectively identifies and characterizes LMs directly from multiphasic CT images. Funding National Natural Science Foundation of China, Guangdong Provincial Key Laboratory of Medical Image Processing.
Collapse
Affiliation(s)
- Hongjie Xin
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Yiwen Zhang
- School of Biomedical Engineering, Southern Medical University, Guangzhou, China
| | - Qianwei Lai
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Naying Liao
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Jing Zhang
- Department of Medical Imaging Center, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Yanping Liu
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
- Department of Gastroenterology, The Second Affiliated Hospital, University of South China, Hengyang, China
| | - Zhihua Chen
- Department of Radiology, The Second Affiliated Hospital, University of South China, Hengyang, China
| | - Pengyuan He
- Department of Infectious Diseases, The Fifth Affiliated Hospital, Sun Yat-sen University, Zhuhai, China
| | - Jian He
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Junwei Liu
- Department of Infectious Diseases and Hepatology Unit, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Yuchen Zhou
- Department of General Surgery, Cancer Center, Integrated Hospital of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China
| | - Wei Yang
- School of Biomedical Engineering, Southern Medical University, Guangzhou, China
| | - Yuanping Zhou
- Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou, China
| |
Collapse
|
2
|
de Araujo EM, Torres US, Racy DJ, Torres LR, Chojniak R, D’Ippolito G. The "streamline phenomenon" of the portal vein flow and its influence on liver involvement by gastrointestinal diseases: current concepts and imaging-based review. Abdom Radiol (NY) 2020; 45:403-415. [PMID: 31768597 DOI: 10.1007/s00261-019-02335-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/25/2022]
Abstract
The streamline flow in the portal system is a phenomenon by which blood from superior mesenteric vein goes preferentially to the right hepatic lobe, while splenic and inferior mesenteric veins divert preferentially to the left lobe. Such a phenomenon results in different patterns of distribution of several liver diseases. The purpose of this article is to discuss the concepts behind the theory of streamline flow and to perform an imaging-based review of representative cases, demonstrating how it may influence the patterns of liver involvement in different gastrointestinal diseases.
Collapse
|
3
|
Sheng W, Dong M, Wang G, Shi X, Gao W, Wang K, Song H, Shi G, Tan X. The diversity between curatively resected pancreatic head and body-tail cancers based on the 8th edition of AJCC staging system: a multicenter cohort study. BMC Cancer 2019; 19:981. [PMID: 31640615 PMCID: PMC6805668 DOI: 10.1186/s12885-019-6178-z] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2019] [Accepted: 09/20/2019] [Indexed: 02/06/2023] Open
Abstract
BACKGROUND To our knowledge, there are no studies to systematically compare the detailed clinical significance between curatively resected pancreatic head (ph) and body-tail (pbt) ductal adenocarcinoma based on the new 8th edition of AJCC staging system (8th AJCC stage) that was just applied in clinical practice in 2018. METHODS Three hundred fifty-one patients with curatively resected pancreatic adenocarcinoma (PC) from three center hospitals were entered into this multicenter cohort study. RESULTS Increasing tumor size (P < 0.001), T stage (T1 + T2 vs T3 + T4, P = 0.003), frequent postoperative liver metastasis (PLM) (P = 0.002) and 8th AJCC stage (IA to VI, P < 0.001; I + II vs III + IV, P = 0.002) were closely associated with the progression of pbt cancers compared with that in ph cancer patients. Moreover, tumor size≥3 cm (P = 0.012), 8th AJCC stage (III + IV) (P = 0.025) and PLM (P = 0.010) were identified as independent risk factors in pbt cancers in logistic analysis. Patients with pbt cancers had a significantly worse overall survival compared with ph cancer patients (P = 0.003). Moreover, pbt was an independent unfavorable factor in multivariate analysis (P = 0.011). In addition to lymph nodes metastasis, 8th AJCC stage, vascular invasion and PLM, increasing tumor size and advanced T stage were also closely associated with the poor prognosis in 131 cases of pbt cancer patients compared with Ph cancer patients. CONCLUSION Pbt, as an independent unfavorable factor for the prognosis of PC patients, are much more aggressive than that in ph cancers according to 8th AJCC staging system. 8th AJCC staging system are more comprehensive and sensitive to reflect the malignant biology of pbt cancers.
Collapse
Affiliation(s)
- Weiwei Sheng
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - Ming Dong
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China.
| | - Guosen Wang
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - Xiaoyang Shi
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - Wei Gao
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - Kewei Wang
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - He Song
- Department of gastrointestinal surgery, the First Hospital, China Medical University, Shenyang, 110001, China
| | - Gang Shi
- Department of general surgery, Cancer hospital of China Medical University, Shenyang, 110042, China
| | - Xiaodong Tan
- Department of thyroid and pancreatic surgery, Shengjing Hospital of China Medical University, Shenyang, 110004, China
| |
Collapse
|
4
|
Khazali AS, Clark AM, Wells A. A Pathway to Personalizing Therapy for Metastases Using Liver-on-a-Chip Platforms. Stem Cell Rev Rep 2017; 13:364-380. [PMID: 28425064 PMCID: PMC5484059 DOI: 10.1007/s12015-017-9735-3] [Citation(s) in RCA: 16] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/06/2023]
Abstract
Metastasis accounts for most cancer-related deaths. The majority of solid cancers, including those of the breast, colorectum, prostate and skin, metastasize at significant levels to the liver due to its hemodynamic as well as tumor permissive microenvironmental properties. As this occurs prior to detection and treatment of the primary tumor, we need to target liver metastases to improve patients' outcomes. Animal models, while proven to be useful in mechanistic studies, do not represent the heterogeneity of human population especially in drug metabolism lack proper human cell-cell interactions, and this gap between animals and humans results in costly and inefficient drug discovery. This underscores the need to accurately model the human liver for disease studies and drug development. Further, the occurrence of liver metastases is influenced by the primary tumor type, sex and race; thus, modeling these specific settings will facilitate the development of personalized/targeted medicine for each specific group. We have adapted such all-human 3D ex vivo hepatic microphysiological system (MPS) (a.k.a. liver-on-a-chip) to investigate human micrometastases. This review focuses on the sources of liver resident cells, especially the iPS cell-derived hepatocytes, and examines some of the advantages and disadvantages of these sources. In addition, this review also examines other potential challenges and limitations in modeling human liver.
Collapse
Affiliation(s)
- A S Khazali
- Department of Pathology, University of Pittsburgh, S711 Scaife Hall, 3550 Terrace St, Pittsburgh, PA, 15261, USA
| | - A M Clark
- Department of Pathology, University of Pittsburgh, S711 Scaife Hall, 3550 Terrace St, Pittsburgh, PA, 15261, USA
| | - A Wells
- Department of Pathology, University of Pittsburgh, S711 Scaife Hall, 3550 Terrace St, Pittsburgh, PA, 15261, USA.
- Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA, USA.
- McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, PA, USA.
- University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA.
- Pittsburgh VA Medical Center, VA Pittsburgh Healthcare System, Pittsburgh, PA, USA.
| |
Collapse
|