Haghbin H, Aziz M. Artificial intelligence and cholangiocarcinoma: Updates and prospects. World J Clin Oncol 2022; 13(2): 125-134 [PMID: 35316928 DOI: 10.5306/wjco.v13.i2.125]
Corresponding Author of This Article
Hossein Haghbin, MD, Doctor, Department of Gastroenterology, Ascension Providence Southfield, 16001 W Nine Mile Road, Southfield, MI 48075, United States. hoshaq@yahoo.com
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Minireviews
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Author contributions: Haghbin H and Aziz M designed and performed the research study. Haghbin H and Aziz M wrote the manuscript; all authors have read and approved the final manuscript.
Conflict-of-interest statement: Authors have no conflict of interest.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Hossein Haghbin, MD, Doctor, Department of Gastroenterology, Ascension Providence Southfield, 16001 W Nine Mile Road, Southfield, MI 48075, United States. hoshaq@yahoo.com
Received: November 9, 2021 Peer-review started: November 9, 2021 First decision: December 27, 2021 Revised: January 9, 2022 Accepted: January 25, 2022 Article in press: January 25, 2022 Published online: February 24, 2022 Processing time: 106 Days and 8.6 Hours
Abstract
Artificial intelligence (AI) is the timeliest field of computer science and attempts to mimic cognitive function of humans to solve problems. In the era of “Big data”, there is an ever-increasing need for AI in all aspects of medicine. Cholangiocarcinoma (CCA) is the second most common primary malignancy of liver that has shown an increase in incidence in the last years. CCA has high mortality as it is diagnosed in later stages that decreases effect of surgery, chemotherapy, and other modalities. With technological advancement there is an immense amount of clinicopathologic, genetic, serologic, histologic, and radiologic data that can be assimilated together by modern AI tools for diagnosis, treatment, and prognosis of CCA. The literature shows that in almost all cases AI models have the capacity to increase accuracy in diagnosis, treatment, and prognosis of CCA. Most studies however are retrospective, and one study failed to show AI benefit in practice. There is immense potential for AI in diagnosis, treatment, and prognosis of CCA however limitations such as relative lack of studies in use by human operators in improvement of survival remains to be seen.
Core Tip: The wide array of modalities available to treat cholangiocarcinoma (CCA) in addition to the diversity of the tumor urges us to use individualized therapy. To establish the proper approach to diagnose, treat, and prognose CCA, analysis of available data is the key to achieve the individualized care. Artificial intelligence can be a potential modality for achieving this goal.
Citation: Haghbin H, Aziz M. Artificial intelligence and cholangiocarcinoma: Updates and prospects. World J Clin Oncol 2022; 13(2): 125-134
The ever-growing rate of technological advancement in medicine has resulted in the era of “Big data”. Artificial intelligence (AI) and its various techniques are used to harness the infinite potential of Big data in medical field[1]. AI, the timeliest field of computer science, involves development of computer algorithms attempting to mimic cognitive function of humans in order to learn and solve problems[2]. Since invention of the first operational computer by Alan Turing in 1940s, we have seen a prodigious rise in AI advancement. Machine learning (ML) is a very practical area of AI that enables computers to learn without direct programming. ML helps machines learn from previous data and improve their learning behavior by gaining experience from data patterns, thereby establishing ever improving predictive models[3]. Various AI techniques including representation learning, natural language processing, and different ML techniques, such as regression trees, support-vector machines (SVM), artificial neural networks (ANN) and more recently, deep learning (DL), have been used in medical field[4]. ML and DL have vastly increased the scope of AI and enabled individualized medicine rather than algorithm-only-based care and has resulted in improved accuracy, efficiency, and outcomes[4].
Despite all the benefits of AI, one should be wary of the drawbacks[5]. The field of AI brings enormous potential however it concurrently brings big ethical problems. ML algorithms, to some extent, function as “black-boxes” where there is difficulty in finding the logic behind the decision by the machine. This will have medicolegal consequences which will be more pronounced as the models become more sophisticated and companies behind ML software reluctant to reveal the details of their software. Moreover, AI poses threats to privacy, data security, and patient autonomy. Lastly, ML algorithms do make mistakes and may not provide accurate results across race, gender, and socioeconomic status spectrum[5].
Cholangiocarcinoma (CCA) is the second most common primary malignancy of the liver. CCA originates from the epithelial cells of the bile ducts exclusive of gallbladder and ampulla of Vater. CCA is an aggressive tumor diagnosed sporadically in advanced stages with high mortality[6]. The incidence of CCA is increasing; therefore, there is increased interest in diagnosis, prognosis, and treatment of this malignancy[7]. Both serum markers and radiologic imaging are used for diagnosis of CCA. A combination of serum markers like liver function tests, carbohydrate antigen (CA) 19-9, and carcinoembryonic antigen (CEA) are utilized to diagnose the disease[8]. The presence of the vast array of serum markers has led to utilization of the markers in novel AI tools in combination with imaging. Positron emission tomography with fluorodeoxyglucose (FDG-PET) integrated with computed tomography (CT) and Magnetic resonance imaging (MRI) in combination with magnetic resonance cholangiopancreatography (MRCP) are valuable tools harnessed by AI to assess the extent of tumor and stage the disease[9,10]. Treatment includes surgical management, neoadjuvant/adjuvant chemotherapy and chemoradiotherapy, hepatic artery radioembolization, and orthotopic liver transplant in selected patients[11-14]. Endoscopic retrograde cholangiopancreatography (ERCP) has two roles of diagnosis and treatment of CCA. Its diagnostic role includes inspecting and providing samples from the biliary system. As palliative treatment, stent placement provides increased quality of life especially in most unresectable cases[15]. Novel AI tools have been able to help in individualizing candidates for each treatment modality.
Increased mortality from CCA in the last decade has coincided with development of AI technology. Figure 1 illustrates how AI can be used to diagnose, treat, and prognose patients with CCA. This review depicts how AI can analyze the radiologic, serologic, and histologic markers of CCA to diagnose, stage, and aid with an individualized treatment plan in addition to giving a prognostic estimate with or without treatment modalities.
Figure 1 Application of artificial intelligence in addressing cholangiocarcinoma.
LR: Logistic regression; SVM: Support-vector machine.
AI has shown promise to aid in diagnosis of CCA. AI is particularly helpful in CCA diagnosis as the condition is not common and there is heterogeneity in anatomical location of the tumor and risk factors of the tumor[16]. This makes the traditional algorithms inferior compared to AI. Many AI tools in the field of ML have been utilized for diagnosis of CCA (Table 1). LR is a linear regression model used for binary classification of problems[17]. SVM is an appropriate model for small samples, high-dimensional, and non-linear patterns assigning labels to objects and has advantage of avoiding “over learning” problem[18]. ANN or multilayer perceptron is an attempt to simulate the biologic nervous system with neurons interconnected able to do parallel processing[17]. Developed by Huang et al, Extreme Learning Machines (ELM) are a type of feedforward neural network models that have shown superiority over SVMs and traditional feedforward neural networks[19]. Convoluted neural network (CNN), a type of DL consists of multilayer of ANN that results in a superior learning ability of complex tasks and has been used in radiology and imaging of the malignancy and associating the radiological data to the clinicopathologic data[20,21]. Every method has their advantages and drawbacks illustrated in Table 1.
Table 1 Advantages and disadvantages of artificial intelligence models used for cholangiocarcinoma diagnosis in radiology.
AI technology
Imaging modalities used in
Advantages
Disadvantages
Logistic regression
US/CT
Interpretable
Low precision
Support-vector machine
US/CT/MRI
Avoids overlearning and dimension disaster problems
Prone to missing data
Extreme learning machine
CT
Does not need high amount of data for training
Slow processing speed
Artificial neural network
CT/MRI
High generalization power
Needs long training time
Convolutional neural network
US/CT/MRI
Higher efficacy and speed as there is no need to compute features as first step
Needs large training data
AI IN THE DIAGNOSIS OF CCA
Serum markers
Evaluation of serum markers is amongst the least invasive and most available data that is present in many patients even before there is a suspicion for diagnosis of CCA. Due to wide availability, these tests are used in adjunct with radiological and other clinical factors in diagnosis of CCA. Sometimes serological models are enough to diagnose the malignancy; for example, Negrini et al[22] developed a ML model that analyzed 15 bile acids of the serum and was able to diagnose CCA with good sensitivity of 79% and excellent accuracy, Area Under Curve (AUC), and specificity of 86.4%, 95%, and 100%, respectively. ANN based model using combination CCA associated carbohydrate antigen and alkaline phosphatase showed promise in diagnosing CCA with a sensitivity and specificity of more than 95%[23].
Cytology
ERCP and Cytology of brushings is a valuable tool for diagnosis of CCA. As a common malignant cause of biliary stricture is CCA, cytology can be crucial in early stages of the malignancy when radiology may have limited roles. Urman et al[24], using a neural network model studying metabolomic and proteomic profile of bile from 36 CCA patients, was able to satisfactorily distinguish CCA from benign stricture with AUC, sensitivity, and specificity of 98.4%, 94.1% and 92.3%, respectively.
Histology
Histology remains the gold standard for diagnosis of malignancies including CCA. From their Shanghai laboratory, Sun et al[25] developed a CNN model for diagnosis of CCA from microscopic hyperspectral pathological slides with promising results. After setting up the first benchmark based on microscopic pathological images consisting of 880 images with pixels manually labeled as tumor or non-tumor for the AI learning, the CNN model was able to diagnose CCA with 88.3% accuracy[25]. AI assistance in histology has not always shown benefits. Stanford University researchers developed an AI diagnostic assistant using DL model to assist pathologists in differentiating hepatocellular carcinoma (HCC) from CCA (26). The model had a good accuracy rate of 84.2% on a set of 80 slides however it failed to improve performance among pathologists [Odds ratio (OR) 1.287, 95%CI: 0.886-1.871]. For all case difficulty levels, the model highly biased the decision of pathologists which led them to wrong diagnosis[26]. The authors concluded that this would question the use of current AI technology for difficult subspecialty tasks[26]. Sometimes CCA can manifest as cancer of unknown primary site (CUP) as it metastasizes to other organs. AI has been used to delineate source of CUP, consisting of 3 to 5% of tumors[27]. CUP-AI-Dx is a CNN model that was trained on more than 18,000 tumors including CCA and has achieved an accuracy of 98.54% in finding the primary site of tumor from the human body system in cross-validation[28].
CT
To elucidate the lesion detected by ultrasound, further workup is required with CT, MRI, and MRCP. As CNN is a DL technique that consists of multilayers of ANN, it has shown great potential especially once it comes to radiology image analysis of pixels. Human yield in diagnosing CCA is limited. Nakai et al[29] have developed CNN models factoring in a combination of CT with serum tumor markers including CEA and CA 19-9. Their CNN model was superior to human radiologists in detecting CCA (0.68 vs 0.45; P = 0.04) [29]. One challenge in diagnosing CCA is differentiating intrahepatic CCA from other intrahepatic malignancies. Xu et al[30] have developed an AI model on 28 intrahepatic lymphomas and 101 CCAs. Their model was able to differentiate between the two tumors with AUC and accuracy both more than 85%. Pannoprat et al[31] have developed CNN model that can differentiate between CCA and hepatocellular carcinoma (the most common primary liver malignancy) with an 88% accuracy. Zhang et al[32] performed a retrospective analysis of contrast enhanced CT of 86 patients with CCA and 46 with combined CCA/HCC tumors, which are difficult to differentiate from CCA necessitating biopsy and surgery. Using ML techniques to classify the lesions as CCA or combined CCA/HCC achieved an AUC of 94.2%[32].
MRI and MRCP
MRI and MRCP have a superior function to diagnose CCA than CT due to ability to illustrate soft tissue, vasculature, and biliary system better than that of CT. ML has been widely utilized in MRI and MRCP. Xu et al[33] and Yu et al[34] each studied MRI of more than 100 patients with CCA and developed SVM models that showed superiority (validation group AUC 87.0% and 90%, respectively). Logeswaran et al[35] in a 2009 study showed 88 to 94% detection rate of Multilayer Perceptron ANN in diagnosis of CCA in MRCP. Yang et al[36] developed an AI model for MRI diagnosis and evaluation of extent of lymph node metastasis of CCA patients. After training the model on 100 CCA patients, the model was able to differentiate high vs low risk CCA groups and lymph node metastasis with AUCs of 80% and 90% in testing cohorts, respectively[36]. Table 2 lists the studies using AI models to diagnose CCA.
Table 2 Studies utilizing artificial intelligence in the diagnosis of cholangiocarcinoma.
Can machine learning radiomics provide pre-operative differentiation of combined hepatocellular cholangiocarcinoma from hepatocellular carcinoma and cholangiocarcinoma to inform optimal treatment planning?
Deep learning for the digital pathologic diagnosis of cholangiocarcinoma and hepatocellular carcinoma: evaluating the impact of a web-based diagnostic assistant
Differentiation of Intrahepatic Cholangiocarcinoma and Hepatic Lymphoma Based on Radiomics and Machine Learning in Contrast-Enhanced Computer Tomography
A Novel Approach to Assessing Differentiation Degree and Lymph Node Metastasis of Extrahepatic Cholangiocarcinoma: Prediction Using a Radiomics-Based Particle Swarm Optimization and Support Vector Machine Model
Automatic Detection and Classification of Focal Liver Lesions Based on Deep Convolutional Neural Networks: A Preliminary Study
Multiphasic CT
CNN
TREATMENT AND PROGNOSIS OF CCA
ML techniques have also been used for treatment and prognosis of CCA. Almost all studies use a combination of radiological, histological, serological, and clinical data for the best results in predicting the survival of the patients and their response to treatment. Table 3 illustrates the studies using AI models to treat and prognose CCA. The fact that such sophisticated models are needed is proof to the complexity of the CCA pathophysiology and ever developing variety of treatment protocols that makes decision making impossible without help of AI technology. One example of such potential is studied by Tsilimigras et al[37]. They constructed a ML model that predicted survival of CCA patients after surgery based on preop serological and radiological data[37]. They conducted an international multi-institutional study on 826 CCA patients, clustering them into groups based on CA 19-9, neutrophil-to-lymphocyte ratio, and tumor size. Their machine learning model showed an excellent agreement with cluster (k = 0.93, 95%CI: 0.90-0.96). This study shows that ML models detect patterns and clusters not detectable to humans using traditional statistical techniques[38]. In this study, AI was able to detect a group of high-risk patients otherwise undetectable. These groups benefit the most from additional neoadjuvant therapy prior to resection as they have a high recurrence[37,38].
Table 3 Studies utilizing artificial intelligence in the treatment and prognostication of cholangiocarcinoma.
The preoperative prognostic value of the radiomics nomogram based on CT combined with machine learning in patients with intrahepatic cholangiocarcinoma
A Novel Classification of Intrahepatic Cholangiocarcinoma Phenotypes Using Machine Learning Techniques: An International Multi-Institutional Analysis
Tumor size, nodal involvement, serology
ML
CT imaging
Another example of tight interrelation between prognosis and treatment is by Jeong et al[39] who elaborated a ML algorithm using the combination of serology, patient characteristics, and CT images of 1421 CCA patients to classify patients to stable and latent risk group. The model was able to predict the disease-free survival between latent and stable groups and response to adjuvant therapy in latent group with excellent ability proven by hazard ratios (HR) of 3.56 and 0.46, respectively (P < 0.001 for both)[39]. Tang et al[40] drew up a predictive model of CCA survival after studying 101 patients with CCA. Their AI model analyzed radiologic characteristics of the CT scan, tumor markers, and past clinical history like cirrhosis with AUC of 78% and 75% for 3-year and 5-year overall survival, respectively[40].
CA 19-9
CA 19-9 as a tumor marker has shown promise in prognosis of CCA. Li et al[41] and Müller et al[42] each validated an AI model to prognosticate the CCA tumors based on clinical, tumor markers such as CA 19-9, serologic like albumin level, and clinical data like nodal metastasis. Li et al[41] model retrospectively studied a total of 1390 patients and achieved a Concordance Index (C-index) superior to the staging system proposed by the 8th edition of the American Joint Committee on Cancer (C-index: 0.693, 95%CI: 0.663-0.723). Müller et al[42] model was able to predict the 1-year survival of patients with an AUC of 89% and 80% for the training and validation sets, respectively.
Palliative measures
Palliative measures like stent placement recommended for inoperable hilar CCAs, are also analyzed by AI models. Shao et al developed an ANN model based on data of 288 CCA patients requiring stent placement that can predict stent occlusion with high AUC of 96% (95%CI: 94-99%)[43].
FUTURE DIRECTIONS
The literature review showed a wealth of studies utilizing AI in CCA, however there is room for much improvement. First, there is need for larger prospective studies including different races, nationalities, and socioeconomic statuses to validate role of AI in diagnosis, treatment, and prognosis of CCA. As study from Stanford showed the AI may not prove to be beneficial in all cases in real life; therefore, in some cases there is need for prospective studies showing AI effectiveness in practice[26]. This precaution is accentuated since there was a lack of negative studies in our review of the literature which can potentially bias toward increased efficacy of AI. Furthermore, the prognostic data should be validated by implementing the data into treatment strategies and seeing an increase in not only survival but also quality of life in CCA patients. One last recommendation for medical field is that healthcare professionals’ education should be improved to prepare them for the ever-increasing role of AI in daily diagnosis, treatment, and prognosis of CCA and at the same time informing them of the current limits and future potentials of the AI technology.
CONCLUSION
In the recent years, we have seen an increase in CCA incidence and, in parallel, a more exponential rise in AI utilization in medicine. AI will be able to utilize the vast amount of data to assist healthcare professionals in addressing CCA. Currently the AI models are showing potential in diagnosis, treatment, and prognosis of CCA. Nonetheless, AI has limits that should be considered; further research is needed to validate use of AI models in real life in use by medical professional to determine their effectiveness and acceptance as auxiliary tools to augment human intelligence. Finally, ethical issues regarding AI including equity and transparency will also need to be addressed to improve acceptance of the technologies by healthcare industry and, more importantly, the patients.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Gu J, Bai J, Shi X, Zhou J, Qiu Y, Wu Y, Jiang C, Sun X, Xu F, Zhang Y, Ding Y. Efficacy and safety of liver transplantation in patients with cholangiocarcinoma: a systematic review and meta-analysis.Int J Cancer. 2012;130:2155-2163.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 19][Cited by in RCA: 21][Article Influence: 1.6][Reference Citation Analysis (0)]
Wu ZF, Zhang HB, Yang N, Zhao WC, Fu Y, Yang GS. Postoperative adjuvant transcatheter arterial chemoembolisation improves survival of intrahepatic cholangiocarcinoma patients with poor prognostic factors: results of a large monocentric series.Eur J Surg Oncol. 2012;38:602-610.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 29][Cited by in RCA: 31][Article Influence: 2.4][Reference Citation Analysis (0)]
Pattanapairoj S, Silsirivanit A, Muisuk K, Seubwai W, Cha'on U, Vaeteewoottacharn K, Sawanyawisuth K, Chetchotsak D, Wongkham S. Improve discrimination power of serum markers for diagnosis of cholangiocarcinoma using data mining-based approach.Clin Biochem. 2015;48:668-673.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 19][Cited by in RCA: 13][Article Influence: 1.3][Reference Citation Analysis (0)]
Urman JM, Herranz JM, Uriarte I, Rullán M, Oyón D, González B, Fernandez-Urién I, Carrascosa J, Bolado F, Zabalza L, Arechederra M, Alvarez-Sola G, Colyn L, Latasa MU, Puchades-Carrasco L, Pineda-Lucena A, Iraburu MJ, Iruarrizaga-Lejarreta M, Alonso C, Sangro B, Purroy A, Gil I, Carmona L, Cubero FJ, Martínez-Chantar ML, Banales JM, Romero MR, Macias RIR, Monte MJ, Marín JJG, Vila JJ, Corrales FJ, Berasain C, Fernández-Barrena MG, Avila MA. Pilot Multi-Omic Analysis of Human Bile from Benign and Malignant Biliary Strictures: A Machine-Learning Approach.Cancers (Basel). 2020;12.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 20][Cited by in RCA: 34][Article Influence: 6.8][Reference Citation Analysis (0)]
Uyumazturk B, Kiani A, Rajpurkar P, Wang A, Ball RL, Gao R, Yu Y, Jones E, Langlotz CP, Martin B, Berry GJ, Ozawa MG, Hazard FK, Brown RA, Chen SB, Wood M, Allard LS, Ylagan L, Ng AY, Shen J. Deep Learning for the Digital Pathologic Diagnosis of Cholangiocarcinoma and Hepatocellular Carcinoma: Evaluating the Impact of a Web-based Diagnostic Assistant.arXiv. 2019;arXiv.
[PubMed] [DOI][Cited in This Article: ]
Zhao Y, Pan Z, Namburi S, Pattison A, Posner A, Balachander S, Paisie CA, Reddi HV, Rueter J, Gill AJ, Fox S, Raghav KPS, Flynn WF, Tothill RW, Li S, Karuturi RKM, George J. CUP-AI-Dx: A tool for inferring cancer tissue of origin and molecular subtype using RNA gene-expression data and artificial intelligence.EBioMedicine. 2020;61:103030.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 55][Cited by in RCA: 56][Article Influence: 11.2][Reference Citation Analysis (0)]
Nakai H, Fujimoto K, Yamashita R, Sato T, Someya Y, Taura K, Isoda H, Nakamoto Y. Convolutional neural network for classifying primary liver cancer based on triple-phase CT and tumor marker information: a pilot study.Jpn J Radiol. 2021;39:690-702.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 3][Cited by in RCA: 23][Article Influence: 5.8][Reference Citation Analysis (0)]
Ponnoprat D, Inkeaw P, Chaijaruwanich J, Traisathit P, Sripan P, Inmutto N, Na Chiangmai W, Pongnikorn D, Chitapanarux I. Classification of hepatocellular carcinoma and intrahepatic cholangiocarcinoma based on multi-phase CT scans.Med Biol Eng Comput. 2020;58:2497-2515.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 10][Cited by in RCA: 23][Article Influence: 4.6][Reference Citation Analysis (0)]
Xu L, Yang P, Liang W, Liu W, Wang W, Luo C, Wang J, Peng Z, Xing L, Huang M, Zheng S, Niu T. A radiomics approach based on support vector machine using MR images for preoperative lymph node status evaluation in intrahepatic cholangiocarcinoma.Theranostics. 2019;9:5374-5385.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 59][Cited by in RCA: 107][Article Influence: 17.8][Reference Citation Analysis (0)]
Yao X, Huang X, Yang C, Hu A, Zhou G, Lei J, Shu J. A Novel Approach to Assessing Differentiation Degree and Lymph Node Metastasis of Extrahepatic Cholangiocarcinoma: Prediction Using a Radiomics-Based Particle Swarm Optimization and Support Vector Machine Model.JMIR Med Inform. 2020;8:e23578.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 12][Cited by in RCA: 11][Article Influence: 2.2][Reference Citation Analysis (0)]
Yang C, Huang M, Li S, Chen J, Yang Y, Qin N, Huang D, Shu J. Radiomics model of magnetic resonance imaging for predicting pathological grading and lymph node metastases of extrahepatic cholangiocarcinoma.Cancer Lett. 2020;470:1-7.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 21][Cited by in RCA: 40][Article Influence: 8.0][Reference Citation Analysis (0)]
Tsilimigras DI, Hyer JM, Paredes AZ, Diaz A, Moris D, Guglielmi A, Aldrighetti L, Weiss M, Bauer TW, Alexandrescu S, Poultsides GA, Maithel SK, Marques HP, Martel G, Pulitano C, Shen F, Soubrane O, Koerkamp BG, Endo I, Pawlik TM. A Novel Classification of Intrahepatic Cholangiocarcinoma Phenotypes Using Machine Learning Techniques: An International Multi-Institutional Analysis.Ann Surg Oncol. 2020;27:5224-5232.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 7][Cited by in RCA: 17][Article Influence: 3.4][Reference Citation Analysis (0)]
Jeong S, Ge Y, Chen J, Gao Q, Luo G, Zheng B, Sha M, Shen F, Cheng Q, Sui C, Liu J, Wang H, Xia Q, Chen L. Latent Risk Intrahepatic Cholangiocarcinoma Susceptible to Adjuvant Treatment After Resection: A Clinical Deep Learning Approach.Front Oncol. 2020;10:143.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 9][Cited by in RCA: 12][Article Influence: 2.4][Reference Citation Analysis (0)]
Müller L, Mähringer-Kunz A, Gairing SJ, Foerster F, Weinmann A, Bartsch F, Heuft LK, Baumgart J, Düber C, Hahn F, Kloeckner R. Survival Prediction in Intrahepatic Cholangiocarcinoma: A Proof of Concept Study Using Artificial Intelligence for Risk Assessment.J Clin Med. 2021;10.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 3][Cited by in RCA: 5][Article Influence: 1.3][Reference Citation Analysis (0)]
Chu H, Liu Z, Liang W, Zhou Q, Zhang Y, Lei K, Tang M, Cao Y, Chen S, Peng S, Kuang M. Radiomics using CT images for preoperative prediction of futile resection in intrahepatic cholangiocarcinoma.Eur Radiol. 2021;31:2368-2376.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 14][Cited by in RCA: 23][Article Influence: 4.6][Reference Citation Analysis (0)]
Ibragimov B, Toesca DAS, Chang DT, Yuan Y, Koong AC, Xing L, Vogelius IR. Deep learning for identification of critical regions associated with toxicities after liver stereotactic body radiation therapy.Med Phys. 2020;47:3721-3731.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 10][Cited by in RCA: 10][Article Influence: 2.0][Reference Citation Analysis (0)]
Liu X, Khalvati F, Namdar K, Fischer S, Lewis S, Taouli B, Haider MA, Jhaveri KS. Can machine learning radiomics provide pre-operative differentiation of combined hepatocellular cholangiocarcinoma from hepatocellular carcinoma and cholangiocarcinoma to inform optimal treatment planning?Eur Radiol. 2021;31:244-255.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 24][Cited by in RCA: 55][Article Influence: 11.0][Reference Citation Analysis (0)]
Midya A, Chakraborty J, Pak L, Zheng J, Jarnagin W, Do RK, Simpson AL.
Deep convolutional neural network for the classification of hepatocellular carcinoma and intrahepatic cholangiocarcinoma. Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis, 2018: 1057528.
[PubMed] [DOI][Cited in This Article: ]
Peng YT, Zhou CY, Lin P, Wen DY, Wang XD, Zhong XZ, Pan DH, Que Q, Li X, Chen L, He Y, Yang H. Preoperative Ultrasound Radiomics Signatures for Noninvasive Evaluation of Biological Characteristics of Intrahepatic Cholangiocarcinoma.Acad Radiol. 2020;27:785-797.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 24][Cited by in RCA: 24][Article Influence: 4.8][Reference Citation Analysis (0)]
Peng Y, Lin P, Wu L, Wan D, Zhao Y, Liang L, Ma X, Qin H, Liu Y, Li X, Wang X, He Y, Yang H. Ultrasound-Based Radiomics Analysis for Preoperatively Predicting Different Histopathological Subtypes of Primary Liver Cancer.Front Oncol. 2020;10:1646.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 22][Cited by in RCA: 40][Article Influence: 8.0][Reference Citation Analysis (0)]
Selvathi D, Malini C, Shanmugavalli P.
Automatic segmentation and classification of liver tumor in CT images using adaptive hybrid technique and Contourlet based ELM classifier. 2013 International Conference on Recent Trends in Information Technology (ICRTIT), 2013: 250-256.
[PubMed] [DOI][Cited in This Article: ]
Wang Q, Wang Z, Sun Y, Zhang X, Li W, Ge Y, Huang X, Liu Y, Chen Y. SCCNN: A Diagnosis Method for Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma Based on Siamese Cross Contrast Neural Network.IEEE Access. 2020;8:85271-85283.
[PubMed] [DOI][Cited in This Article: ]
Wang CJ, Hamm CA, Savic LJ, Ferrante M, Schobert I, Schlachter T, Lin M, Weinreb JC, Duncan JS, Chapiro J, Letzen B. Deep learning for liver tumor diagnosis part II: convolutional neural network interpretation using radiologic imaging features.Eur Radiol. 2019;29:3348-3357.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 55][Cited by in RCA: 82][Article Influence: 13.7][Reference Citation Analysis (0)]
Zhou J, Wang W, Lei B, Ge W, Huang Y, Zhang L, Yan Y, Zhou D, Ding Y, Wu J. Automatic Detection and Classification of Focal Liver Lesions Based on Deep Convolutional Neural Networks: A Preliminary Study.Front Oncol. 2020;10:581210.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 8][Cited by in RCA: 25][Article Influence: 6.3][Reference Citation Analysis (0)]