Detection Of Liver Tumors Research Articles

Real-Time Liver Tumor Detection with a Multi-Class Ensemble Deep Learning Framework

Real-Time Liver Tumor Detection with a Multi-Class Ensemble Deep Learning Framework

DESMPA-Deep CNN: Double Exponential Smoothing Marine Predator Algorithm-Based Deep CNN for Liver Tumor Segmentation and Classification

Nowadays, liver cancers are well-known cancers attacking the human body. The larger segment of liver carcinomas is highly affected by alcohol-associated hepatitis as well as cirrhosis circumstances. In order to increase the survival rate of humans, premature and early diagnosis of liver cancer is very vital. Currently, discriminating the tumor and liver segments from medical imaging with the help of entirely automated computer-aided software is a highly difficult task, because the liver tumor may differ from one human being to another human being. Thus, this paper endeavors to present a new liver tumor segmentation and classification approach by incorporating an optimization-enabled deep learning model. Here, the adopted model comprises four stages, pre-processing, liver segmentation, liver tumor detection, and classification. To begin with, the collected images are given to the pre-processing stage using the African vulture’s optimization-based adaptive histogram equalization (AVOAHE). The integration of the African vulture’s optimization (AVO) concept in adaptive histogram equalization is named AVOAHE. After the image pre-processing phase, the liver region is segmented by utilizing kernel-based fuzzy C-means clustering. Then, the detection of liver tumor is performed using the Swin-Unet. Finally, tumor detection is done by utilizing the deep convolutional neural network (Deep CNN). Moreover, a hybrid optimization approach, called double exponential smoothing (DES)-marine predator algorithm (MPA), is developed by combining DES and MPA, which is used to train Deep CNN and serves as an enhancement. Finally, the experimental analysis has shown that the adopted technique has achieved better results in terms of accuracy, specificity, sensitivity, and precision with values of 0.957, 0.969, 0.941 and 0.915, respectively.

An Enhanced Approach to Intelligent Computer-Assisted Localization of Liver Tumor on Computed Tomography Images

Computed Tomography (CT) Imaging is frequently used to find liver cancer. CT imaging of the liver generates cross-sectional images of the abdominal region. The task of segmenting a liver tumor on CT images is tedious due to the anatomic complexity of the liver. Therefore, liver tumor detection has been performed manually or semi-automatically which is an exhaustive process. Further, a tumor may be present on only some slices of the CT scan and hence there is a need to automatically identify images that contain tumors from acquired CT images and then locate a region of the tumor on those images for further diagnosis. Automatic detection helps the radiologist to obtain fast, accurate, and effective results. Accordingly, in this paper, an enhanced approach is proposed based on deep learning, efficient fuzzy c-means (EFCM) clustering, and a region-based algorithm for the detection of a liver tumor. A deep learning model built on a convolutional neural network is used to identify whether the CT image contains a tumor. If it contains the tumor, then the proposed EFCM and region-based algorithms are applied to locate the tumor on the CT image. The EFCM algorithm is combined with a region-based technique to automatically determine the initial pixel and threshold values required to segment the liver tumor. The proposed method is applied to several CT abdominal images. The results of the tumor segmentation are validated by comparing them with the tumor regions detected by radiologists. The estimated results of the experiment show that the proposed technique successfully detects the liver tumor on the CT image with less than 2% relative error.

Leveraging radiomics and AI for precision diagnosis and prognostication of liver malignancies.

Liver tumors, whether primary or metastatic, have emerged as a growing concern with substantial global health implications. Timely identification and characterization of liver tumors are pivotal factors in order to provide optimum treatment. Imaging is a crucial part of the detection of liver tumors; however, conventional imaging has shortcomings in the proper characterization of these tumors which leads to the need for tissue biopsy. Artificial intelligence (AI) and radiomics have recently emerged as investigational opportunities with the potential to enhance the detection and characterization of liver lesions. These advancements offer opportunities for better diagnostic accuracy, prognostication, and thereby improving patient care. In particular, these techniques have the potential to predict the histopathology, genotype, and immunophenotype of tumors based on imaging data, hence providing guidance for personalized treatment of such tumors. In this review, we outline the progression and potential of AI in the field of liver oncology imaging, specifically emphasizing manual radiomic techniques and deep learning-based representations. We discuss how these tools can aid in clinical decision-making challenges. These challenges encompass a broad range of tasks, from prognosticating patient outcomes, differentiating benign treatment-related factors and actual disease progression, recognizing uncommon response patterns, and even predicting the genetic and molecular characteristics of the tumors. Lastly, we discuss the pitfalls, technical limitations and future direction of these AI-based techniques.

Novel utilization and quantification of Xsight diaphragm tracking for respiratory motion compensation in Cyberknife Synchrony treatment of liver tumors.

The Xsight lung tracking system (XLTS) utilizes an advanced image processing algorithm to precisely identify the position of a tumor and determine its location in orthogonal x-ray images, instead of finding fiducials, thereby minimizing the risk of fiducial insertion-related side effects. To assess and gauge the effectiveness of CyberKnife Synchrony in treating liver tumors located in close proximity to or within the diaphragm, we employed the Xsight diaphragm tracking system (XDTS), which was based on the XLTS. We looked back at the treatment logs of 11 patients (8/11 [XDTS], 3/11 [Fiducial-based Target Tracking System-FTTS]) who had liver tumors in close proximity to or within the diaphragm. And the results are compared with the patients who undergo the treatment of FTTS. The breathing data information was calculated as a rolling average to reduce the effect of irregular breathing. We tested the tracking accuracy with a dynamic phantom (18023-A) on the basis of patient-specific respiratory curve. The average values for the XDTS and FTTS correlation errors were 1.38±0.65 versus 1.50±0.26mm (superior-inferior), 1.28±0.48 versus 0.40±0.09mm (left-right), and 0.96±0.32 versus 0.47±0.10mm(anterior-posterior), respectively. The prediction errors for two methods of 0.65±0.16 versus 5.48±3.33mm in the S-I direction, 0.34±0.10 versus 1.41±0.76mm in the A-P direction, and 0.22±0.072 versus 1.22±0.48mm in the L-R direction. The coverage rate of FTTS slightly less than that of XDTS, such as 96.53±8.19% (FTTS) versus 98.03±1.54 (XDTS). The prediction error, the motion amplitude, and the variation of the respiratory center phase were strongly related to each other. Especially, the higher the amplitude and the variation, the higher the prediction error. The diaphragm has the potential to serve as an alternative to gold fiducial markers for detecting liver tumors in close proximity or within it. We also found that we needed to reduce the motion amplitude and train the respiration of the patients during liver radiotherapy, as well as control and evaluate their breathing.

Accurate artificial intelligence method for abnormality detection of CT liver images

Liver cancer is a leading cause of death worldwide and poses a significant challenge to physicians in terms of accurate diagnosis and treatment. AI-powered segmentation and classification algorithms can play a vital role in assisting physicians in detecting and diagnosing liver tumors. However, liver tumor classification is a difficult task due to factors such as noise, non-homogeneity, and significant appearance variations in cancerous tissue. In this study, we propose a novel approach to automatically segmenting and classifying liver tumors. Our proposed framework comprises three main components: a preprocessing unit to enhance picture contrast, a Masked Recurrent Convolutional Neural Network (RCNN) for liver segmentation, and a pixel-wise classification unit for identifying abnormalities in the liver. When our models are applied to the challenging MICCAI’2027 liver tumor segmentation (LITS) database, we achieve Dice similarity coefficients of 96% and 98% for liver segmentation and lesion identification, respectively. We also demonstrate the efficiency of our proposed framework by comparing it with similar strategies for tumor segmentations. The proposed approach achieved high accuracy, sensitivity, specificity, and F1 score parameters for liver segmentation and lesion identification. These results were evaluated using the Dice similarity coefficient and compared with similar strategies for tumor segmentation. Our approach holds promise for improving the accuracy and speed of liver tumor detection and diagnosis, which could have significant implications for patient outcomes.

Transformer dense center network for liver tumor detection

The detection of small-scale lesions is a challenging task that plays an important role in the diagnosis, treatment, and prognosis of liver disease. In recent years, the one-stage anchor-free object detection network has achieved significant results in liver tumor detection. However, there is a drawback in this network that it has weak ability of capturing global contextual information and detecting small-scale liver tumors. For this purpose, we propose a liver tumor detection method named TDCenterNet by designing a dense residual attention CenterNet network and Transformer structure block. The main framework of TDCenterNet is depicted as follows. Firstly, a novelty dense residual attention CenterNet network is designed as a feature extraction network, in which dense connections are embedded between the residual structure blocks to strengthen the feature flow between different residual structure blocks to improve identification ability. Secondly, to capture the global context information of the lesion, the multi-head attention layer of the Transformer, i.e., Transformer conversion and recognition block, is imposed to a dense residual attention CenterNet network by replacing the 3 × 3 spatial convolutions of Convert Block and Identity Block in the last layer of the dense residual attention CenterNet network. Then, the Atrous spatial pyramid pooling is integrated to the second convert block and identity block to further enhance the multiscale expression ability of capturing lesions with different sizes. Finally, knowledge distillation is performed to train TDCerneterNet to improve detection performance of TDCenterNet. Experiments have shown that the performance of TDCenterNet is significantly higher than comparison methods.

Single preoperative application of indocyanine green for liver function assessment and better intraoperative detection of liver tumors

Single preoperative application of indocyanine green for liver function assessment and better intraoperative detection of liver tumors

[Applications] 9. Development of an AI System for Liver Tumor Detection from Abdominal Ultrasound Images

[Applications] 9. Development of an AI System for Liver Tumor Detection from Abdominal Ultrasound Images

Translational Insights into Molecular Mechanisms of Chemical Hepatocarcinogenesis for Improved Human Risk Assessment

Background: Hepatocellular carcinoma (HCC) is a prevalent liver cancer with major risk factors being hepatitis viral infections, alcohol, non-alcoholic fatty liver disease, and aflatoxin exposure. Both genotoxic and non-genotoxic agents can induce HCC through mechanisms involving DNA damage, oxidative stress, chronic inflammation, and disrupted signaling pathways like MAPK/ERK, PI3K/AKT, WNT/β-catenin and PPARα. While rodent assays are utilized to detect potential chemical hepatocarcinogens, species differences in pathways like PPARα and CAR/PXR activation impact human risk assessment. Purpose: This analysis provides an updated, critical examination of species concordance in mechanisms of hepatic carcinogenesis to inform human safety assessment of rodent liver tumor findings. Main Body: Rodent assays including 2-year bioassays, transgenic models, and short-term studies detect liver tumors through lifetime exposure or early biomarkers. However, rodent-specific PPARα and CAR/PXR activation, along with human risk factors like hepatitis, highlight key interspecies differences. Determining mode of action relevance requires evaluating mechanistic validity and pivotal key events leading to tumors across species. Non-genotoxic compounds eliciting rodent liver tumors can activate PPARα, CAR/PXR, and other pathways triggering increased cell replication; but downstream signaling may differ in human liver. Understanding applicability of these mechanisms in humans as well as incorporating human risk factors into experimental models is critical for accurate risk assessment. Conclusion: In summary, elucidating conserved versus divergent molecular mechanisms of hepatic carcinogenesis between rodents and humans is essential for appropriately interpreting rodent findings and safeguarding human health through science-based risk assessment frameworks and regulatory decision-making processes around potential chemical hazards.

Convolutional neural network-based classifiers for liver tumor detection using computed tomography scans

Convolutional neural network-based classifiers for liver tumor detection using computed tomography scans

Automatic liver tumor detection and classification using the hyper tangent fuzzy C-Means and improved fuzzy SVM

Automatic liver tumor detection and classification using the hyper tangent fuzzy C-Means and improved fuzzy SVM

Circulating microRNA Panels for Detection of Liver Cancers and Liver-Metastasizing Primary Cancers.

Malignant liver tumors, including primary malignant liver tumors and liver metastases, are among the most frequent malignancies worldwide. The disease carries a poor prognosis and poor overall survival, particularly in cases involving liver metastases. Consequently, the early detection and precise differentiation of malignant liver tumors are of paramount importance for making informed decisions regarding patient treatment. Significant research efforts are currently directed towards the development of diagnostic tools for different types of cancer using minimally invasive techniques. A prominent area of focus within this research is the evaluation of circulating microRNA, for which dysregulated expression is well documented in different cancers. Combining microRNAs in panels using serum or plasma samples derived from blood holds great promise for better sensitivity and specificity for detection of certain types of cancer.

Spatiotemporal knowledge teacher–student reinforcement learning to detect liver tumors without contrast agents

Detecting Liver tumors without contrast agents (CAs) has shown great potential to advance liver cancer screening. It enables the provision of a reliable liver tumor-detecting result from non-enhanced MR images comparable to the radiologists’ results from CA-enhanced MR images, thus eliminating the high risk of CAs, preventing an experience gap between radiologists and simplifying clinical workflows. In this paper, we proposed a novel spatiotemporal knowledge teacher–student reinforcement learning (SKT-RL) as a safe, speedy, and inexpensive contrast-free technology for liver tumor detection. Our SKT-RL builds a teacher–student framework to realize the exploring of explicit liver tumor knowledge from a teacher network on clear contrast-enhanced images to guide a student network to detect tumors from non-enhanced images directly. Importantly, our STK-RL enables three novelties in aspects of construction, transferring, and optimization to tumor knowledge to improve the guide effect. (1) A new spatiotemporal ternary knowledge set enables the construction of accurate knowledge that allows understanding of DRL’s behavior (what to do) and reason (why to do it) behind reliable detection within each state and between their related historical states. (2) A novel pixel momentum transferring strategy enables detailed and controlled knowledge transfer ability. It transfers knowledge at a pixel level to enlarge the explorable space of transferring and control how much knowledge is transferred to prevent over-rely of the student to the teacher. (3) A phase-trend reward function designs different evaluations according to different detection phases to optimal for each phase in high-precision but also allows reward trend to constraint the evaluation to improve stability. Comprehensive experiments on a generalized liver tumor dataset with 375 patients (including hemangiomas, hepatocellular carcinoma, and normal controls) show that our novel SKT-RL attains a new state-of-the-art performance (improved precision by at least 4% when comparing the six recent advanced methods) in the task of liver tumor detection without CAs. The results proved that our SKT-DRL has greatly promoted the development and deployment of contrast-free liver tumor technology.

Slice-Fusion: Reducing False Positives in Liver Tumor Detection for Mask R-CNN.

Automatic liver tumor detection from computed tomography (CT) makes clinical examinations more accurate. However, deep learning-based detection algorithms are characterized by high sensitivity and low precision, which hinders diagnosis given that false-positive tumors must first be identified and excluded. These false positives arise because detection models incorrectly identify partial volume artifacts as lesions, which in turn stems from the inability to learn the perihepatic structure from a global perspective. To overcome this limitation, we propose a novel slice-fusion method in which mining the global structural relationship between the tissues in the target CT slices and fusing the features of adjacent slices according to the importance of the tissues. Furthermore, we design a new network based on our slice-fusion method and Mask R-CNN detection model, called Pinpoint-Net. We evaluated proposed model on the Liver Tumor Segmentation Challenge (LiTS) dataset and our liver metastases dataset. Experiments demonstrated that our slice-fusion method not only enhance tumor detection ability via reducing the number of false-positive tumors smaller than 10mm, but also improve segmentation performance. Without bells and whistles, a single Pinpoint-Net showed outstanding performance in liver tumor detection and segmentation on LiTS test dataset compared with other state-of-the-art models.

Dual-path Network for Liver and Tumor Segmentation in CT Images Using Swin Transformer Encoding Approach.

Liver and tumor segmentation from CT images is a complex and crucial step in achieving full-course adaptive radiotherapy and also plays an essential role in computer-aided clinical diagnosis systems. Deep learning-based methods play an important role in achieving automatic segmentation. This research aims to improve liver tumor detection performance by proposing a dual path feature extracting strategy and employing Swin-Transformer. The hierarchical Swin-Transformer is embedded into the encoder and decoder and combined with CNN to form a dual coding path structure incorporating long-range dependencies and multi-scale contextual connections to capture coarse-tuned features at different semantic scales. The features of the two encoding paths and the upsampling path are fused, tested and validated with LITS and in-house datasets. The proposed method has a DG of 97.95% and a DC of 96.2% for liver segmentation; a DG of 80.6% and a DC of 68.1% for tumor segmentation; and a classification study of the tumor dataset shows a DG of 91.1% and a DC of 87.2% for large and continuous tumors and a DG of 71.7% and a DC of 66.4% for small and scattered tumors. Swin-Transformer can be used as a robust encoder for medical image segmentation networks and, combined with CNN networks, can better recover local spatial information and enhance feature representation. Accurate localization before segmentation can achieve better results for small and scattered tumors.

A Coarse-to-Fine Fusion Network for Small Liver Tumor Detection and Segmentation: A Real-World Study.

Liver tumor semantic segmentation is a crucial task in medical image analysis that requires multiple MRI modalities. This paper proposes a novel coarse-to-fine fusion segmentation approach to detect and segment small liver tumors of various sizes. To enhance the segmentation accuracy of small liver tumors, the method incorporates a detection module and a CSR (convolution-SE-residual) module, which includes a convolution block, an SE (squeeze and excitation) module, and a residual module for fine segmentation. The proposed method demonstrates superior performance compared to conventional single-stage end-to-end networks. A private liver MRI dataset comprising 218 patients with a total of 3605 tumors, including 3273 tumors smaller than 3.0 cm, were collected for the proposed method. There are five types of liver tumors identified in this dataset: hepatocellular carcinoma (HCC); metastases of the liver; cholangiocarcinoma (ICC); hepatic cyst; and liver hemangioma. The results indicate that the proposed method outperforms the single segmentation networks 3D UNet and nnU-Net as well as the fusion networks of 3D UNet and nnU-Net with nnDetection. The proposed architecture was evaluated on a test set of 44 images, with an average Dice similarity coefficient (DSC) and recall of 86.9% and 86.7%, respectively, which is a 1% improvement compared to the comparison method. More importantly, compared to existing methods, our proposed approach demonstrates state-of-the-art performance in segmenting small objects with sizes smaller than 10 mm, achieving a Dice score of 85.3% and a malignancy detection rate of 87.5%.

RETRACTED: Deep learning based two-fold segmentation model for liver tumor detection

This article has been retracted. A retraction notice can be found at https://doi.org/10.3233/JIFS-219433.

6,6'-2 H2 ] fructose as a deuterium metabolic imaging probe in liver cancer.

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths. Imaging plays a crucial role in the early detection of HCC, although current methods are limited in their ability to characterize liver lesions. Most recently, deuterium metabolic imaging (DMI) has been demonstrated as a powerful technique for the imaging of metabolism in vivo. Here, we assess the metabolic flux of [6,6'-2 H2 ] fructose in cell cultures and in subcutaneous mouse models at 9.4 T. We compare these rates with the most widely used DMI probe, [6,6'-2 H2 ] glucose, exploring the possibility of developing 2 H fructose to overcome the limitations of glucose as a novel DMI probe for detecting liver tumors. Comparison of the in vitro metabolic rates implies their similar glycolytic metabolism in the TCA cycle due to comparable production rates of 2 H glutamate/glutamine (glx) for the two precursors, but overall higher glycolytic metabolism from 2 H glucose because of a higher production rate of 2 H lactate. In vivo kinetic studies suggest that HDO can serve as a robust reporter for the consumption of the precursors in liver tumors. As fructose is predominantly metabolized in the liver, deuterated water (HDO) produced from 2 H fructose is probably less contaminated from whole-body metabolism in comparison with glucose. Moreover, in studies of the normal liver, 2 H fructose is readily converted to 2 H glx, enabling the characterization of 2 H fructose kinetics. This overcomes a major limitation of previous 2 H glucose studies in the liver, which were unable to confidently discern metabolic flux due to overlapped signals of 2 H glucose and its metabolic product, 2 H glycogen. This suggests a unique role for 2 H fructose metabolism in HCC and the normal liver, making it a useful approach for assessing liver-related diseases and the progression to oncogenesis.

Influence of grey wolf optimization feature selection on gradient boosting machine learning techniques for accurate detection of liver tumor

Malignant growth in liver results in liver tumor. The most common types of liver cancer are primary liver disease and secondary liver disease. Most malignant growths are benign tumors, and the condition they cause, essential liver disease, is the end result. Cancer of the liver is a potentially fatal disease that can only be cured by combining a number of different treatments. Machine learning, feature selection and image processing have the capability to provide a framework for the accurate detection of liver diseases. The processing of images is one of the components that come together to form this group. When utilized for the purpose of reviewing previously recorded visual information, the instrument performs at its highest level of effectiveness. The importance of feature selection on machine learning algorithms for the early and accurate diagnosis of liver tumors is discussed in this article. The input consists of images from a CT scan of the liver. These images are preprocessed by discrete wavelet transform. Discrete wavelet transforms increase resolution by compressing the images. Images are segmented in parts to identify region of interest by K Means algorithm. Features are selected by grey wolf optimization technique. Classification is performed by Gradient boosting, support vector machine and random forest. GWO Gradient boosting is performing better in accurate classification and prediction of liver cancer.