Ontology-Driven Approach for Liver MRI Classification and HCC Detection

2021 ◽  
Vol 35 (12) ◽  
Author(s):  
Rim Messaoudi ◽  
Faouzi Jaziri ◽  
Achraf Mtibaa ◽  
Faïez Gargouri ◽  
Antoine Vacavant
Keyword(s):  
Medical Image ◽  
Arterial Phase ◽  
Deep Convolutional Neural Networks ◽  
Real Patient ◽  
Low Contrast ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
Medical Image Classification ◽  
Magnetic Resonance Imaging Mri ◽  
Clinical Problems

Reading and interpreting the medical image still remains the most challenging task in radiology. Through the important achievement of deep Convolutional Neural Networks (CNN) in the context of medical image classification, various clinical applications have been provided to detect lesions from Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scans. In the diagnosis process for the liver cancer from Dynamic Contrast-Enhanced MRI (DCE-MRI), radiologists consider three phases during contrast injection: before injection, arterial phase, and portal phase for instance. Even if the contrast agent helps in enhancing the tumoral tissues, the diagnosis may be very difficult due to the possible low contrast and pathological tissues surrounding the tumors (cirrhosis). Alongside, in the medical field, ontologies have proven their effectiveness to solve several clinical problems such as offering shareable terminologies, vocabularies, and databases. In this article, we propose a multi-label CNN classification approach based on a parallel preprocessing algorithm. This algorithm is an extension of our previous work cited in the International Conference on Pattern Recognition and Artificial Intelligence (ICPRAI) 2020. The aim of our approach is to ameliorate the detection of HCC lesions and to extract more information about the detected tumor such as the stage, the localization, the size, and the type thanks to the use of ontologies. Moreover, the integration of such information has improved the detection process. In fact, experiments conducted by testing with real patient cases have shown that the proposed approach reached an accuracy of 93% using MRI patches of [Formula: see text] pixels, which is an improvement compared with our previous works.

Differentiation of intrapancreatic accessory spleen from small hypervascular neuroendocrine tumor of the pancreas: textural analysis on contrast-enhanced computed tomography

2018 ◽  
Vol 60 (5) ◽  
pp. 553-560 ◽  
Author(s):  
Xubo Lin ◽  
Lei Xu ◽  
Aiqin Wu ◽  
Chuangen Guo ◽  
Xiao Chen ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Lymph Node ◽  
Roc Analysis ◽  
Texture Features ◽  
Arterial Phase ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
Lymph Node Enlargement ◽  
Enhanced Computed Tomography ◽  
Contrast Enhanced Computed Tomography

Background Intrapancreatic accessory spleens (IPASs) are usually misdiagnosed as pancreatic neuroendocrine tumors (PNETs). Texture analysis is valuable in tumor detection, diagnosis, and staging. Purpose To identify the potential of texture features in differentiating IPASs from small hypervascular PNETs. Material and Methods Twenty-one patients with PNETs and 13 individuals with IPASs who underwent pretreatment dynamic contrast-enhanced computed tomography (CT) were retrospectively analyzed. The routine imaging features—such as location, size, margin, cystic or solid appearance, enhancement degree and pattern, and lymph node enlargement—were recorded. Texture features, such as entropy, skewness, kurtosis, and uniformity, on contrast-enhanced images were analyzed. Receiver operating characteristic (ROC) analysis was performed to differentiate IPASs from PNETs. Results No significant differences were observed in margin, enhancement degree (arterial and portal phase), lymph node enlargement, or size between PNETs and IPASs (all P > 0.05). However, IPASs usually showed heterogeneous enhancement at the arterial phase and the same degree of enhancement as the spleen at the portal phase, both of which were greater than those of PNETs (69% vs. 35%, P = 0.06; 100% vs. 29%, P = 0.04). Entropy and uniformity were significantly different between IPASs and PNETs at moderate (1.5) and high sigma values (2.5) (both P < 0.01). ROC analysis showed that uniformity at moderate and high sigma had the highest area under the curve (0.82 and 0.89) with better sensitivity (85.0–95.0%) and acceptable specificity (75.0–83.3%) for differentiating IPASs from PNETs. Conclusions Texture parameters have potential in differentiating IPASs from PNETs.

scholarly journals Characteristics of hepatic solitary necrotic nodules on contrast-enhanced ultrasonography

2021 ◽  
Author(s):  
Chunyu Lu ◽  
Shaoshan Tang ◽  
Xiaoyue Zhang ◽  
Yang Wang ◽  
Kaiming Wang ◽  
...  
Keyword(s):  
The Other ◽  
Arterial Phase ◽  
Focal Lesions ◽  
Multiple Lesions ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
The Us ◽  
Contrast Enhanced Ultrasonography ◽  
Dynamic Enhancement ◽  
Three Phases

Abstract Background:To summarize the characteristics of solitary necrotic nodules (SNN) in the liver observed under contrast-enhanced ultrasonography (CEUS).Methods:Conventional ultrasonography (US) and CEUS were performed in 24 patients who were confirmed to have SNN by pathological assessment. The US data and dynamic enhancement patterns of CEUS were recorded and retrospectively analyzed.Results:Ten of 24 patients underwent surgical resection, while the other 14 patients underwent a puncture biopsy to be confirmed as SNN. Among the 24 patients, 13 patients had a single lesion and 11 patients had multiple lesions. The largest lesion was selected for CEUS examination for patients with multiple lesions. Eleven patients presented no enhancement in all three phases, while the other 13 patients presented with a peripheral thin rim-like enhancement in the arterial phase, an iso-enhancement in the portal phase and delayed phase. However, no enhancement in the interior of the lesions was detected during three phases of CEUS.Conclusions:SNN has characteristic findings on the CEUS, which play an important role in the differential diagnoses of liver focal lesions.

scholarly journals Characteristics of hepatic solitary necrotic nodules on contrast-enhanced ultrasonography

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Chunyu Lu ◽  
Shaoshan Tang ◽  
Xiaoyue Zhang ◽  
Yang Wang ◽  
Kaiming Wang ◽  
...  
Keyword(s):  
The Other ◽  
Arterial Phase ◽  
Focal Lesions ◽  
Multiple Lesions ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
The Us ◽  
Contrast Enhanced Ultrasonography ◽  
Dynamic Enhancement ◽  
Three Phases

Abstract Background To summarize the characteristics of solitary necrotic nodules (SNN) in the liver observed under contrast-enhanced ultrasonography (CEUS). Methods Conventional ultrasonography (US) and CEUS were performed in 24 patients who were confirmed to have SNN by pathological assessment. The US data and dynamic enhancement patterns of CEUS were recorded and retrospectively analyzed. Results Ten of 24 patients underwent surgical resection, while the other 14 patients underwent a puncture biopsy to be confirmed as SNN. Among the 24 patients, 13 patients had a single lesion and 11 patients had multiple lesions. The largest lesion was selected for CEUS examination for patients with multiple lesions. Eleven patients presented no enhancement in all three phases, while the other 13 patients presented with a peripheral thin rim-like enhancement in the arterial phase, an iso-enhancement in the portal phase and delayed phase. However, no enhancement in the interior of the lesions was detected during three phases of CEUS. Conclusions SNN has characteristic findings on the CEUS, which play an important role in the differential diagnoses of liver focal lesions.

Medical Image Classification

2019 ◽  
pp. 59-80
Author(s):  
Jesu Vedha Nayahi J. ◽  
Gokulakrishnan K.
Keyword(s):  
Image Classification ◽  
Medical Image ◽  
Clinical Symptoms ◽  
Input Image ◽  
Vital Role ◽  
Disease Classification ◽  
Medical Image Classification ◽  
Scanned Images ◽  
Magnetic Resonance Imaging Mri ◽  
Severity Of The Disease

Diagnosis of diseases at the right stage with optimal accuracy is a significant requirement in the medical field. Apart from diagnosis from clinical symptoms, diagnosis based on scanned images of both internal and external organs is playing a vital role in understanding the severity of the disease. Classification is a field of study derived from artificial intelligence, and today it is widely used in medical image classification. These techniques are used to classify the different stages of a disease or different variant diseases possible in an organ from different types of input images such as magnetic resonance imaging (MRI), computed tomography (CT), x-ray, fundus images, iris images, etc. Various preprocessing techniques are used to select the relevant features from the input image to form the feature set. The classifiers are trained using the feature set to generate models. The generated models can be optimized to improve the performance. Various metrics such as accuracy, coverage, precision, recall, and FMeasure are used to measure the accuracy.

scholarly journals Mass-forming cholangiocarcinoma

2019 ◽  
Vol 2 (1) ◽  
pp. 15-17
Author(s):  
Tanita Suttichaimongkol ◽  
Kawin Tangvoraphonkchai ◽  
Arin Pisanuwongse
Keyword(s):  
Diagnostic Tools ◽  
Arterial Phase ◽  
Case Presentation ◽  
Intraductal Growth ◽  
Portal Venous Phase ◽  
Contrast Enhanced ◽  
Liver Cancers ◽  
Contrast Magnetic Resonance ◽  
Venous Phase ◽  
Magnetic Resonance Imaging Mri

Cholangiocarcinoma is the second most common primary liver cancers. It is arising from epithelial cells of the biliary tract. It has been categorized to intrahepatic and extrahepatic. The Intrahepatic orperipheral cholangiocarcinoma can be presented as mass-forming, periductal infiltrating and intraductal growth. Many patients of mass-forming cholangiocarcinoma have symptoms such as abdominal pain about 85% but some patients don’t have any symptoms. This is the difficult cancer to diagnose. While patients were having any symptom, the disease was an advanced stage (unresectable). The diagnostic tools for assess this disease are imaging modalities include ultrasound (US), computed tomography (CT) with contrast, magnetic resonance imaging (MRI) with contrast. However, the goal standard for confirm diagnosis is tissue pathology. This article showed a case presentation and reviewed the imaging appearance of mass-forming cholangiocarcinoma.   Figure 1  Axial non-contrast (A), axial contrast enhanced in arterial phase (B), axial contrast enhanced in portal venous phase (C) and axial contrast enhanced in 5-minute delay phase (D) CT scans show a large ill-defined hypodense mass at hepatic segment 7/8, about 7.0x7.0x5.0 cm in APxLxH diameter, which has poor enhancement on arterial phase with gradual progressive enhancement on portal venous and 5-minute delay phase. Coronal contrast enhanced in portal venous phase CT scan (E) shows mass confined in peripheral area of right hepatic lobe with hepatic vein abutment.

scholarly journals Characteristics of hepatic solitary necrotic nodules on contrast-enhanced ultrasonography

2020 ◽  
Author(s):  
Chunyu Lu ◽  
Shaoshan Tang ◽  
Xiaoyue Zhang ◽  
Yang Wang ◽  
Kaiming Wang ◽  
...  
Keyword(s):  
Differential Diagnoses ◽  
Arterial Phase ◽  
Focal Lesions ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
The Us ◽  
Contrast Enhanced Ultrasonography ◽  
Dynamic Enhancement ◽  
Delayed Phase ◽  
Single Lesion

Abstract Background:To summarize the characteristics of solitary necrotic nodules (SNN) in the liver observed under contrast-enhanced ultrasonography (CEUS), and to improve the differential diagnoses value of CEUS on liver focal lesions.Methods:Conventional ultrasonography (US) and CEUS were performed in 24 patients who were confirmed to have SNN by pathological assessment. The US data and dynamic enhancement patterns of CEUS were recorded and retrospectively analyzed.Results:All patients underwent CEUS, and of these, 10 patients underwent surgical resection, while 14 patients underwent a puncture biopsy to confirm SNN. Among the 24 patients, 13 patients had a single lesion and 11 patients had multiple lesions. Eleven patients presented no enhancement in all three phases, while the other 13 patients presented with a peripheral thin rim-like enhancement in the arterial phase, an iso-enhancement in the portal phase and delayed phase, and no enhancement in the interior of the lesions. Conclusions:SNN has characteristic findings on the CEUS, which play an important role in the differential diagnoses of liver focal lesions.

Construction of Deep Convolutional Neural Networks For Medical Image Classification

2019 ◽  
Vol 9 (2) ◽  
pp. 1-15 ◽  
Author(s):  
Rama A ◽  
Kumaravel A ◽  
Nalini C
Keyword(s):  
Image Classification ◽  
Medical Image ◽  
High Performance ◽  
Fine Tuning ◽  
Heterogeneous Environments ◽  
Feature Maps ◽  
Deep Convolutional Neural Networks ◽  
Convolutional Network ◽  
Final Layer ◽  
Medical Image Classification

Implementing image processing tools demands its components produce better results in critical applications like medical image classification. TensorFlow is one open source with a machine learning framework for high performance and operates in heterogeneous environments. It heralds broad attention at a fine tuning of parameters for obtaining the final models, to obtain better performance. The main aim of this article is to prove the appropriate steps for the classification techniques for diagnosing the diseases with better accuracy. The proposed convolutional network is comprised of three convolutional layers, preceded by average pooling with a size equal to the size of the final feature maps. The final layer in this network has two outputs, corresponding to the number of classes considered to be either normal or abnormal. To train and evaluate such networks like the Deep Convolutional Neural Network (DCNN), a dataset of 2000 x-ray images of lungs was used and a comparative analysis between the proposed DCNN against previous methods is also made.

scholarly journals Characteristics of hepatic solitary necrotic nodules on contrast-enhanced ultrasonography

2020 ◽  
Author(s):  
Chunyu Lu ◽  
Shaoshan Tang ◽  
Xiaoyue Zhang ◽  
Yang Wang ◽  
Kaiming Wang ◽  
...  
Keyword(s):  
Arterial Phase ◽  
Focal Lesions ◽  
Portal Phase ◽  
Contrast Enhanced ◽  
The Us ◽  
Puncture Biopsy ◽  
Contrast Enhanced Ultrasonography ◽  
Dynamic Enhancement ◽  
Delayed Phase ◽  
Single Lesion

Abstract Background:To summarize the characteristics of solitary necrotic nodules (SNN) in the liver observed under contrast-enhanced ultrasonography (CEUS).Methods:Conventional ultrasonography (US) and CEUS were performed in 24 patients who were confirmed to have SNN by pathological assessment. The US data and dynamic enhancement patterns of CEUS were recorded and retrospectively analyzed.Results:All patients underwent CEUS, and of these, 10 patients underwent surgical resection, while 14 patients underwent a puncture biopsy to confirm SNN. Among the 24 patients, 13 patients had a single lesion and 11 patients had multiple lesions，the largest lesion was selected for CEUS examination. Eleven patients presented no enhancement in all three phases, while the other 13 patients presented with a peripheral thin rim-like enhancement in the arterial phase, an iso-enhancement in the portal phase and delayed phase, and no enhancement in the interior of the lesions. Conclusions:SNN has characteristic findings on the CEUS, which play an important role in the differential diagnoses of liver focal lesions.

scholarly journals A rotation and translation invariant method for 3D organ image classification using deep convolutional neural networks

2019 ◽  
Vol 5 ◽  
pp. e181 ◽  
Author(s):  
Kh Tohidul Islam ◽  
Sudanthi Wijewickrema ◽  
Stephen O’Leary
Keyword(s):  
Image Classification ◽  
Medical Image ◽  
Three Dimensional ◽  
Disease Diagnosis ◽  
3D Image ◽  
Patient Orientation ◽  
Deep Convolutional Neural Networks ◽  
Translation Invariant ◽  
Medical Image Classification ◽  
Image Modality

Three-dimensional (3D) medical image classification is useful in applications such as disease diagnosis and content-based medical image retrieval. It is a challenging task due to several reasons. First, image intensity values are vastly different depending on the image modality. Second, intensity values within the same image modality may vary depending on the imaging machine and artifacts may also be introduced in the imaging process. Third, processing 3D data requires high computational power. In recent years, significant research has been conducted in the field of 3D medical image classification. However, most of these make assumptions about patient orientation and imaging direction to simplify the problem and/or work with the full 3D images. As such, they perform poorly when these assumptions are not met. In this paper, we propose a method of classification for 3D organ images that is rotation and translation invariant. To this end, we extract a representative two-dimensional (2D) slice along the plane of best symmetry from the 3D image. We then use this slice to represent the 3D image and use a 20-layer deep convolutional neural network (DCNN) to perform the classification task. We show experimentally, using multi-modal data, that our method is comparable to existing methods when the assumptions of patient orientation and viewing direction are met. Notably, it shows similarly high accuracy even when these assumptions are violated, where other methods fail. We also explore how this method can be used with other DCNN models as well as conventional classification approaches.

Sign in / Sign up

Export Citation Format

Share Document