Computer auxiliary diagnosis technique of detecting cholangiocarcinoma based on medical imaging: A review

doi:10.1016/j.cmpb.2021.106265

Computer Methods and Programs in Biomedicine

Volume 208, September 2021, 106265

https://doi.org/10.1016/j.cmpb.2021.106265 Get rights and content

Highlights

•
A comprehensive review of medical imaging-based CAD systems using AI techniques for cholangiocarcinoma detecting and diagnosis.
•
Summarized the complete workflow of a CAD system.
•
Explored and listed specific techniques of each steps of CAD for CCA based on CT, MR and US imaging.
•
Optimization of image processing, and efficiency research of AI algorithms and medical application were suggested to develop precise CAD system.

Abstract

Background and objectives

Cholangiocarcinoma (CCA) is one of the most aggressive human malignant tumors and is becoming one of the main factors of death and disability globally. Specifically, 60% to 70% of CCA patients were diagnosed with local invasion or distant metastasis and lost the chance of radical operation. The overall median survival time was less than 12 months. As a non-invasive diagnostic technology, medical imaging consisting of computed tomography (CT) imaging, magnetic resonance imaging (MRI), and ultrasound (US) imaging, is the most effectively and commonly used method to detect CCA. The computer auxiliary diagnosis (CAD) system based on medical imaging is helpful for rapid diagnosis and provides credible “second opinion” for specialists. The purpose of this review is to categorize and review the CAD technique of detecting CCA based on medical imaging.

Methods

This work applies a four-level screening process to choose suitable publications. 125 research papers published in different academic research databases were selected and analyzed according to specific criteria. From the five steps of medical image acquisition, processing, analysis, understanding and verification of CAD combined with artificial intelligence algorithms, we obtain the most advanced insights related to CCA detection.

Results

This work provides a comprehensive analysis and comparison analysis of the current CAD systems of detecting CCA. After careful investigation, we find that the main detection methods are traditional machine learning method and deep learning method. For the detection, the most commonly used method is semi-automatic segmentation algorithm combined with support vector machine classifier method, combination of which has good detection performance. The end-to-end training mode makes deep learning method more and more popular in CAD systems. However, due to the limited medical training data, the accuracy of deep learning method is unsatisfactory.

Conclusions

Based on analysis of artificial intelligence methods applied in CCA, this work is expected to be truly applied in clinical practice in the future to improve the level of clinical diagnosis and treatment of it. This work concludes by providing a prediction of future trends, which will be of great significance for researchers in the medical imaging of CCA and artificial intelligence.

Graphical abstract

Section snippets

1. Introduction

As a common tumor, cholangiocarcinoma (CCA) originates from epithelial cells of the bile duct and has a high degree of malignancy [1]. Its location can range anywhere from the capillary bile duct to the common bile duct. As a malignant tumor second only to hepatocellular carcinoma (HCC), there has been a dramatic increase in its incidence rate in the past 40 years [2]. However, in contrast to the biological behavior of HCC, CCA is more likely to infiltrate into the bile duct wall and invade the

2. Publications search strategy and organization of the review

In this review, in order to search the publications relevant to CAD technique of detecting CCA based on medical imaging, we referred research publication databases consisting of Springer^TM, IEEEXplore^TM, Pubmed^TM, ScienceDirect^TM, Google Scholar^TM, and WANFANG DATA^TM. These papers were screened against the background of medical imaging findings and artificial intelligence methods.

We applied a four-level screening process to determine the most suitable publications for this review. In the first

3. Imaging of cholangiocarcinoma

In clinical practice, the early-stage diagnosis of CCA is usually completed by medical imaging. CT scanning is the main modality for diagnosis and staging of CCA [17]. MRI can provide a detailed anatomical description of any vascular involvement of the bile duct [18]. Although US cannot be used for accurate staging, it plays an important role in early screening [19].

4. CAD of cholangiocarcinoma

In medical images, the output of a CAD system is obtained by quantitative analysis of the characteristics of relevant image data. Its function is to help radiologists improve the accuracy of diagnosis and the consistency of image and disease interpretation. Although its results can only be used as an auxiliary means, and diseases cannot be completely diagnosed by it, a CAD system can improve the accuracy of doctors' diagnosis. The reason is that, firstly, in traditional diagnostic methods, the

5. Discussion and future research

Medical imaging is increasingly used as an auxiliary method for the detection of CCA. However, currently, the application of AI technology in the auxiliary diagnosis of CCA has made some progress, but there remains a few shortcomings and limitations. In this section, we not only consider the pitfalls of existed CAD methods using medical images for detecting CCA but provide some potential directions to make CAD systems work better for further research.

There are three major subfields that can be

Conclusions

In this paper, we reviewed CAD techniques of detecting CCA based on medical imaging, including CT, MR, and US. We introduced algorithms and techniques used in image preprocessing, feature extraction and selection, classification, and performance estimation of a CAD system. Moreover, further improvement in CAD systems of CCA has a specific scope. We have provided a prediction of future trends, which will be of great significance for researchers in the medical imaging of both CCA and AI.

Declaration of Competing Interest

We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.

Acknowledgments

This work was supported by the Shanghai Natural Science Foundation of China [Grant No. 19ZR1421500].

References (125)

S. Yadav et al.
Neoadjuvant vs. adjuvant chemotherapy for cholangiocarcinoma: A propensity score matched analysis
Eur. J. Surg. Oncol.
(2019)
P. Lambin et al.
Radiomics: extracting more information from medical images using advanced feature analysis
Eur. J. Cancer
(2012)
H. Zhang et al.
Multi-source transfer learning via multi-kernel support vector machine plus for B-mode ultrasound-based computer-aided diagnosis of liver cancers
IEEE J. Biomed. Health Inform.
(2021)
C. Xu et al.
Synthesis of gadolinium-enhanced liver tumors on nonenhanced liver MR images using pixel-level graph reinforcement learning
Med. Image Anal.
(2021)
D.T. Kushnure et al.
MS-UNet: a multi-scale UNet with feature recalibration approach for automatic liver and tumor segmentation in CT images
Comput. Med. Imaging Graph.
(2021)
J.W. Valle et al.
Biliary tract cancer
Lancet
(2021)
T.P. Coroller et al.
CT-based radiomic signature predicts distant metastasis in lung adenocarcinoma
Radiother. Oncol.
(2015)
Y.W. Liu et al.
A novel method for accurate extraction of liver capsule and auxiliary diagnosis of liver cirrhosis based on high-frequency ultrasound images
Comput. Biol. Med.
(2020)
E.C. Lo et al.
Hepatocellular carcinoma and intrahepatic cholangiocarcinoma: imaging for diagnosis, tumor response to treatment and liver response to radiation
Semin. Radiat. Oncol.
(2018)
H.-X. Xu et al.
Enhancement pattern of hilar cholangiocarcinoma: contrast-enhanced ultrasound versus contrast-enhanced computed tomography
Eur. J. Radiol.
(2010)

I.R. Valente et al.

Automatic 3D pulmonary nodule detection in CT images: a survey

Comput. Methods Progr. Biomed.

(2016)

M. Kosmin et al.

Rapid advances in auto-segmentation of organs at risk and target volumes in head and neck cancer

Radiother. Oncol.

(2019)

D. Shao et al.

Characteristic matching-based adaptive fast bilateral filter for ultrasound speckle reduction

Pattern Recognit. Lett.

(2013)

A. Wong et al.

Monte Carlo despeckling of transrectal ultrasound images of the prostate

Digit. Signal Process.

(2012)

Y. Zhang et al.

Fractional subpixel diffusion and fuzzy logic approach for ultrasound speckle reduction

Pattern Recognit.

(2010)

I. Elamvazuthi et al.

Despeckling of ultrasound images of bone fracture using multiple filtering algorithms

Math. Comput. Model.

(2013)

S. Tsantis et al.

Inter-scale wavelet analysis for speckle reduction in thyroid ultrasound images

Comput. Med. Imaging Graph.

(2007)

S. Ghose et al.

A survey of prostate segmentation methodologies in ultrasound, magnetic resonance and computed tomography images

Comput. Methods Progr. Biomed.

(2012)

L.D. Cohen

On active contour models and balloons

CVGIP: Image Understanding

(1991)

L. Huang et al.

Treatment progress of cholangiocarcinoma

J. Hepatopancreatobiliary Surg.

(2012)

S.K. Saha et al.

Forty-year trends in cholangiocarcinoma incidence in the U.S.: intrahepatic disease on the rise

Oncologist

(2016)

A.I. Tobkes et al.

Liver biopsy: review of methodology and complications

Dig. Dis.

(1995)

C. Vicas et al.

Usefulness of textural analysis as a tool for noninvasive liver fibrosis staging

J. Med. Ultrason.

(2011)

G.S. Lodwick

Computer-aided diagnosis in radiology. A research plan

Invest. Radiol.

(1966)

X.L. Chen et al.

Research progress of cholangiocarcinoma

Chin. J. Clin. Exp. Pathol.

(2013)

I. Endo et al.

Intrahepatic cholangiocarcinoma: rising frequency, improved survival, and determinants of outcome after resection

Ann. Surg.

(2008)

L. Wu et al.

Trends in the incidence, treatment and outcomes of patients with intrahepatic cholangiocarcinoma in the USA: facility type is associated with margin status, use of lymphadenectomy and overall survival

World J. Surg.

(2019)

D.I. Tsilimigras et al.

A machine-based approach to preoperatively identify patients with the most and least benefit associated with resection for intrahepatic cholangiocarcinoma: an international multi-institutional analysis of 1146 patients

Ann. Surg. Oncol.

(2020)

Liu Chung-Ming et al.

Computer-aided diagnosis of liver tumors on computed tomography images

Comput. Methods Progr. Biomed.

(2017)

C.M. Yang et al.

Cholangiocarcinoma evaluation via imaging and artificial intelligence

Oncology

(2020)

Y. Zhu et al.

Value of contrast-enhanced CT texture analysis in predicting IDH mutation status of intrahepatic cholangiocarcinoma

Sci. Rep.

(2021)

B. Hma et al.

Quantitative analysis of gadoxetic acid-enhanced MRI for the differential diagnosis of focal liver lesions: comparison between estimated intralesional gadoxetic acid retention by T1 mapping and conventional processing methods

Eur. J. Radiol.

(2021)

Z.S. Chen et al.

Research advances in the clinical diagnosis of intrahepatic cholangiocarcinoma

J. Clin. Hepatol.

(2019)

M.G. Lee et al.

MR cholangiopancreatography of pancreaticobiliary diseases: comparing single-shot RARE and multislice HASTE sequences

AJR Am. J. Roentgenol.

(1998)

A. Justus et al.

Non-breath-hold magnetic resonance cholangiography-preliminary results and review of literature

Indian J. Radiol. Imaging

(1999)

D.I. Tsilimigras et al.

ASO author reflections: use of machine learning to identify patients with intrahepatic cholangiocarcinoma who could benefit more from neoadjuvant therapies

Ann. Surg. Oncol.

(2020)

J.H. Min et al.

Intrahepatic mass-forming cholangiocarcinoma: arterial enhancement patterns at MRI and prognosis

Radiology

(2019)

R.J. Gillies et al.

Radiomics: images are more than pictures, they are data

Radiology

(2016)

A. Wibmer et al.

Haralick texture analysis of prostate MRI: utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different gleason scores

Eur. Radiol.

(2015)

Y. Lin et al.

Gray-scale ultrasound-based radiomics in distinguishing hepatocellular carcinoma from intrahepatic mass-forming Chinese

J. Med. Imaging

(2020)

X. Liu et al.

Extracting and describing liver capsule contour in high-frequency ultrasound image for early HBV cirrhosis diagnosis

W. Shuo Hong et al.

Learning to diagnose cirrhosis via combined liver capsule and parenchyma ultrasound image features

X. Liu et al.

Learning to diagnose cirrhosis with liver capsule guided ultrasound image classification

Sensors

(2017)

X. Liu et al.

Computer-aided cirrhosis diagnosis via automatic liver capsule extraction and combined geometry-texture features

Z. Xu et al.

Diagnosis of cirrhosis stage via deep neural network

J.W. Zhao et al.

Early diagnosis of cirrhosis via automatic location and geometric description of liver capsule

Vis. Comput.

(2018)

G.X. Fu et al.

A liver capsule extraction algorithm based on high frequency ultrasound images

Electron. Sci. Technol.

(2019)

H.F. Wang et al.

Advances in clinical diagnosis and treatment of distal cholangiocarcinoma

Henan Med. Res.

(2015)

P. Sungkasubun et al.

Ultrasound screening for cholangiocarcinoma could detect premalignant lesions and early-stage diseases with survival benefits: a population-based prospective study of 4225 subjects in an endemic area

BMC Cancer

(2016)

I.S. Oliveira et al.

Cholangiocarcinoma: classification, diagnosis, staging, imaging features, and management

Abdom. Radiol.

(2017)

Cited by (11)

Research related to the diagnosis of prostate cancer based on machine learning medical images: A review
2024, International Journal of Medical Informatics
Prostate cancer is currently the second most prevalent cancer among men. Accurate diagnosis of prostate cancer can provide effective treatment for patients and greatly reduce mortality. The current medical imaging tools for screening prostate cancer are mainly MRI, CT and ultrasound. In the past 20 years, these medical imaging methods have made great progress with machine learning, especially the rise of deep learning has led to a wider application of artificial intelligence in the use of image-assisted diagnosis of prostate cancer.
This review collected medical image processing methods, prostate and prostate cancer on MR images, CT images, and ultrasound images through search engines such as web of science, PubMed, and Google Scholar, including image pre-processing methods, segmentation of prostate gland on medical images, registration between prostate gland on different modal images, detection of prostate cancer lesions on the prostate.
Through these collated papers, it is found that the current research on the diagnosis and staging of prostate cancer using machine learning and deep learning is in its infancy, and most of the existing studies are on the diagnosis of prostate cancer and classification of lesions, and the accuracy is low, with the best results having an accuracy of less than 0.95. There are fewer studies on staging. The research is mainly focused on MR images and much less on CT images, ultrasound images.
Machine learning and deep learning combined with medical imaging have a broad application prospect for the diagnosis and staging of prostate cancer, but the research in this area still has more room for development.
Upregulation of circ_0059961 suppresses cholangiocarcinoma development by modulating miR-629-5p/SFRP2 axis
2022, Pathology Research and Practice
Citation Excerpt :
CCA is the second greatest frequency among hepatobiliary malignancies, accounting for 15–20% of primary hepatobiliary tumors [2]. The prognosis for CCA is extremely bad, with a median survival time of less than three years, and the incidence is growing particularly in western countries [3]. Previous studies have investigated the complex pathogenesis of cholangiocarcinoma in several aspects such as inflammatory response, cholestasis, genetic and chromosomal factors, and dysregulation of genes expression, but no effective treatment or remission for this condition has been obtained so far, and there was a high recurrence rate of CCA even with surgical removal [4].
Cholangiocarcinoma (CCA) is a dangerous malignancy with a poor prognosis due to inefficient chemotherapy and surgery, and its pathophysiology could be linked to circular RNA (circRNAs) dysregulation. As a result, we wanted to see what role circ_0059961 plays in CCA.
The quantitative real-time polymerase chain reaction (qRT-PCR) was used to detect the expression of circ_0059961, miR-629–5p, and secreted frizzled related protein 2 (SFRP2). Cell counting kit-8 (CCK-8) and 5-Ethynyl-2′-deoxyuridine (Edu) assay were used to determine the role of circ_0059961 in proliferation. The MMP (Δψm) assay was used to assess cell apoptosis. Transwell assay was used to detect cell migration and invasion. The SFRP2 protein and epithelia-mesenchymal transition (EMT) process related proteins expression were measured using Western blot. The putative relationship between miR-629–5p and circ_0059961 or SFRP2 was validated by dual-luciferase reporter assay. The circ_0059961 roles in cholangiocarcinoma were also investigated by tumor xenograft assay.
Circ_0059961 expression was decreased in CCA tissues and carcinoma cells. Overexpressed circ_0059961 reduced tumor cell proliferation, migration, and invasion while inducing apoptosis. Circ_0059961 was proven to be a target of miR-629–5p, while miR-629–5p was confirmed to be a target of SFRP2. Circ_0059961 targeted miR-629–5p to modulate SFRP2 expression. Upregulation of miR-629–5p reversed the tumor-suppressive effects of circ_0059961 overexpression in rescue trials. Furthermore, SFRP2 overexpression restored miR-629–5p enrichment-promoted cell proliferation, migration, and invasion. Upregulated circ_0059961 reduced solid tumor growth in vivo.
Upregulation of circ_0059961 augmented SFRP2 expression by targeting miR-629–5p, which blocked cholangiocarcinoma tumor cell proliferation, migration, and invasion.
A multilayer network-enabled ultrasonic image series analysis approach for online cancer drug delivery monitoring
2022, Computer Methods and Programs in Biomedicine
Citation Excerpt :
According to the literatures, morphologic features and texture features [14–16] were widely used in the ultrasonic image analysis. The well-known image feature extractions approaches were investigated, such as the statistics of pixel values within the region of interest (ROI) [2–5], the histogram of oriented gradients (HOG) [17], and local binary pattern (LBP) [18]. Furthermore, machine learning-based approaches are also developed rapidly for ultrasonic image analysis in recent years [19,20].
The objective of this study is to develop an effective data-driven methodology for the online monitoring of cancer drug delivery guided by the ultrasonic images. To achieve this goal, effective image quantification and accurate feature extraction play a critical role on image-guided drug delivery (IGDD) monitoring. However, the existing image-guided approaches in such area are mainly focused on the analysis for individual images rather than the image series. In fact, the temporal patterns between consecutive images may contain critical information and it is necessary to be considered in the monitoring analysis. In addition, the conventional approaches, such as the pure intensity-based method, also do not sufficiently consider the effects of noise in the ultrasonic images, which also limits the monitoring sensitivity and accuracy. To address the challenges, this paper proposed a novel multilayer network-enabled IGDD (MNE-IGDD) monitoring approach. The contributions of the proposed method can be summarized into three aspects: (1) formulate the sequential ultrasound images to a multilayer network by the proposed spatial-regularized distance; (2) detect drug delivery area based on community detection algorithm of multilayer network; and (3) quantify the drug delivery progress by incorporating the image intensity-based features with the detected community. Both the detected communities and feature increment percentages are applied as the evaluation metric for validation. A simulation study was conducted and this method was also applied to a real-world mouse colon tumor treatment case study under three temperature conditions. Both simulation and the real-world case studies demonstrated that the proposed method is promising to achieve satisfactory monitoring performance in clinical trials.
Applications of artificial intelligence in biliary tract cancers
2024, Indian Journal of Gastroenterology
Artificial intelligence in medical imaging for cholangiocarcinoma diagnosis: A systematic review with scientometric analysis
2023, Journal of Gastroenterology and Hepatology (Australia)
Attention-guided salient object detection using autoencoder regularization
2023, Applied Intelligence

View all citing articles on Scopus

View full text

Computer auxiliary diagnosis technique of detecting cholangiocarcinoma based on medical imaging: A review

Highlights

Abstract

Background and objectives

Methods

Results

Conclusions

Graphical abstract

Section snippets

1. Introduction

2. Publications search strategy and organization of the review

3. Imaging of cholangiocarcinoma

4. CAD of cholangiocarcinoma

5. Discussion and future research

Conclusions

Declaration of Competing Interest

Acknowledgments

Eur. J. Surg. Oncol.

Eur. J. Cancer

IEEE J. Biomed. Health Inform.

Med. Image Anal.

Comput. Med. Imaging Graph.

Lancet

Radiother. Oncol.

Comput. Biol. Med.

Semin. Radiat. Oncol.

Eur. J. Radiol.

Comput. Methods Progr. Biomed.

Radiother. Oncol.

Pattern Recognit. Lett.

Digit. Signal Process.

Pattern Recognit.

Math. Comput. Model.

Comput. Med. Imaging Graph.

Comput. Methods Progr. Biomed.

CVGIP: Image Understanding

Treatment progress of cholangiocarcinoma

J. Hepatopancreatobiliary Surg.

Forty-year trends in cholangiocarcinoma incidence in the U.S.: intrahepatic disease on the rise

Oncologist

Liver biopsy: review of methodology and complications

Dig. Dis.

Usefulness of textural analysis as a tool for noninvasive liver fibrosis staging

J. Med. Ultrason.

Computer-aided diagnosis in radiology. A research plan

Invest. Radiol.

Research progress of cholangiocarcinoma

Chin. J. Clin. Exp. Pathol.

Intrahepatic cholangiocarcinoma: rising frequency, improved survival, and determinants of outcome after resection

Ann. Surg.

Trends in the incidence, treatment and outcomes of patients with intrahepatic cholangiocarcinoma in the USA: facility type is associated with margin status, use of lymphadenectomy and overall survival

World J. Surg.

A machine-based approach to preoperatively identify patients with the most and least benefit associated with resection for intrahepatic cholangiocarcinoma: an international multi-institutional analysis of 1146 patients

Ann. Surg. Oncol.

Computer-aided diagnosis of liver tumors on computed tomography images

Comput. Methods Progr. Biomed.

Cholangiocarcinoma evaluation via imaging and artificial intelligence

Oncology

Value of contrast-enhanced CT texture analysis in predicting IDH mutation status of intrahepatic cholangiocarcinoma

Sci. Rep.

Quantitative analysis of gadoxetic acid-enhanced MRI for the differential diagnosis of focal liver lesions: comparison between estimated intralesional gadoxetic acid retention by T1 mapping and conventional processing methods

Eur. J. Radiol.

Research advances in the clinical diagnosis of intrahepatic cholangiocarcinoma

J. Clin. Hepatol.

MR cholangiopancreatography of pancreaticobiliary diseases: comparing single-shot RARE and multislice HASTE sequences

AJR Am. J. Roentgenol.

Non-breath-hold magnetic resonance cholangiography-preliminary results and review of literature

Indian J. Radiol. Imaging

ASO author reflections: use of machine learning to identify patients with intrahepatic cholangiocarcinoma who could benefit more from neoadjuvant therapies

Ann. Surg. Oncol.

Intrahepatic mass-forming cholangiocarcinoma: arterial enhancement patterns at MRI and prognosis

Radiology

Radiomics: images are more than pictures, they are data

Radiology

Haralick texture analysis of prostate MRI: utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different gleason scores

Eur. Radiol.

Gray-scale ultrasound-based radiomics in distinguishing hepatocellular carcinoma from intrahepatic mass-forming Chinese

J. Med. Imaging

Extracting and describing liver capsule contour in high-frequency ultrasound image for early HBV cirrhosis diagnosis

Learning to diagnose cirrhosis via combined liver capsule and parenchyma ultrasound image features