Effective segmentation and classification of tumor on liver MRI and CT images using multi-kernel K-means clustering

Artificial intelligence in liver imaging: methods and applications

Liver disease is regarded as one of the major health threats to humans. Radiographic assessments hold promise in terms of addressing the current demands for precisely diagnosing and treating liver diseases, and artificial intelligence (AI), which excels at automatically making quantitative assessments of complex medical image characteristics, has made great strides regarding the qualitative interpretation of medical imaging by clinicians. Here, we review the current state of medical-imaging-based AI methodologies and their applications concerning the management of liver diseases. We summarize the representative AI methodologies in liver imaging with focusing on deep learning, and illustrate their promising clinical applications across the spectrum of precise liver disease detection, diagnosis and treatment. We also address the current challenges and future perspectives of AI in liver imaging, with an emphasis on feature interpretability, multimodal data integration and multicenter study. Taken together, it is revealed that AI methodologies, together with the large volume of available medical image data, might impact the future of liver disease care.

Radiomics: a primer on high-throughput image phenotyping

Radiomics is a high-throughput approach to image phenotyping. It uses computer algorithms to extract and analyze a large number of quantitative features from radiological images. These radiomic features collectively describe unique patterns that can serve as digital fingerprints of disease. They may also capture imaging characteristics that are difficult or impossible to characterize by the human eye. The rapid development of this field is motivated by systems biology, facilitated by data analytics, and powered by artificial intelligence. Here, as part of Abdominal Radiology's special issue on Quantitative Imaging, we provide an introduction to the field of radiomics. The technique is formally introduced as an advanced application of data analytics, with illustrating examples in abdominal radiology. Artificial intelligence is then presented as the main driving force of radiomics, and common techniques are defined and briefly compared. The complete step-by-step process of radiomic phenotyping is then broken down into five key phases. Potential pitfalls of each phase are highlighted, and recommendations are provided to reduce sources of variation, non-reproducibility, and error associated with radiomics.

K-Means Clustering Algorithm–Based Functional Magnetic Resonance for Evaluation of Regular Hemodialysis on Brain Function of Patients with End-Stage Renal Disease

This research was to evaluate the effects of regular hemodialysis (HD) on the brain function of patients with end-stage renal disease (ESRD). Resting-state functional magnetic resonance imaging (rs-fMRI) based on improved k-means clustering algorithm (k-means) was proposed to scan the brains of 30 regular dialysis patients with end-stage renal disease (ESRD) (experimental group) and 30 normal volunteers (control group). The proposed algorithm was compared with the traditional k-means algorithm and mean shift algorithm and applied to the magnetic resonance scan of patients with ESRD on long-term regular HD. The results showed that the neuropsychological cognitive function (NSCF) evaluation result of the test group was much better than that of the control group, and the difference was statistically obvious (P < 0.05). The results of blood biochemistry, Digit Symbol Substitution Test (DSST), and Montreal Cognitive Assessment Scale (MoCA) in the test group showed no statistical difference compared with those in the control group. The running time of the improved k-means algorithm was dramatically shorter than that of traditional k-means algorithm, showing statistical difference (P < 0.05). Comparison among the improved and traditional k-means algorithm and mean shift algorithm suggested that the improved k-means algorithm showed a lower error rate for image segmentation, and the differences were statistically remarkable (P < 0.05). In conclusion, the improved k-means algorithm showed better time efficiency and the lowest error rate in processing rs-fMRI images than the traditional k-means algorithm and mean shift algorithm, and the effects of regular HD on the brains of patients with ESRD were evaluated effectively.

Diagnosis of Liver Tumors in Human CT Images Based on the LiverNet Approach

Liver cancer contributes to the increasing mortality rate in the world. Therefore, early detection may lead to a decrease in morbidity and increase the chance of survival rate. This research offers a computer-aided diagnosis system, which uses computed tomography scans to categorize hepatic tumors as benign or malignant. The 3D segmented liver from the LiTS17 dataset is passed through a Convolutional Neural Network (CNN) to detect and classify the existing tumors as benign or malignant. In this work, we propose a novel light CNN with eight layers and just one conventional layer to classify the segmented liver. This proposed model is utilized in two different tracks; the first track uses deep learning classification and achieves a 95.6% accuracy. Meanwhile, the second track uses the automatically extracted features together with a Support Vector Machine (SVM) classifier and achieves 100% accuracy. The proposed network is light, fast, reliable, and accurate. It can be exploited by an oncological specialist, which will make the diagnosis a simple task. Furthermore, the proposed network achieves high accuracy without the curation of images, which will reduce time and cost.

Dual-Energy Computed Tomography for Evaluation of Breast Cancer Follow-Ups: Comparison of Virtual Monoenergetic Images and Iodine-Map

Differentiating tumor tissue from dense breast tissue can be difficult. Dual-energy CT (DECT) could be suitable for making diagnoses at breast cancer follow-ups. This study investigated the contrast in DECT images and iodine maps for patients with breast cancer being followed-up. Chest CT images captured in 2019 were collected. Five cases of metastatic breast cancer in the lungs were analyzed; the contrast-to-noise ratio (for breast tissue and muscle: CNRb and CNRm, respectively), tumor-to-breast mammary gland ratio (T/B), and tumor-to-muscle ratio (T/M) were calculated. For 84 cases of no metastasis, monochromatic spectral and iodine maps were obtained to compare differences under various breast densities using the K-means algorithm. The optimal T/B, T/M, and CNRb (related to mammary glands) were achieved for the 40-keV image. Conversely, CNRm (related to lungs) was better for higher energy. The optimal balance was achieved at 80 keV. T/B, T/M, and CNR were excellent for iodine maps, particularly for density > 25%. In conclusion, energy of 80 keV is the parameter most suitable for observing the breast and lungs simultaneously by using monochromatic spectral images. Adding iodine mapping can be appropriate when a patient’s breast density is greater than 25%.