Hepatocellular Carcinoma Recognition from Ultrasound Images Using Combinations of Conventional and Deep Learning Techniques.

Hepatocellular Carcinoma (HCC) is the most frequent malignant liver tumor and the third cause of cancer-related deaths worldwide. For many years, the golden standard for HCC diagnosis has been the needle biopsy, which is invasive and carries risks. Computerized methods are due to achieve a noninvasive, accurate HCC detection process based on medical images. We developed image analysis and recognition methods to perform automatic and computer-aided diagnosis of HCC. Conventional approaches that combined advanced texture analysis, mainly based on Generalized Co-occurrence Matrices (GCM) with traditional classifiers, as well as deep learning approaches based on Convolutional Neural Networks (CNN) and Stacked Denoising Autoencoders (SAE), were involved in our research. The best accuracy of 91% was achieved for B-mode ultrasound images through CNN by our research group. In this work, we combined the classical approaches with CNN techniques, within B-mode ultrasound images. The combination was performed at the classifier level. The CNN features obtained at the output of various convolution layers were combined with powerful textural features, then supervised classifiers were employed. The experiments were conducted on two datasets, acquired with different ultrasound machines. The best performance, above 98%, overpassed our previous results, as well as representative state-of-the-art results.

Hepatocellular Carcinoma Automatic Diagnosis within CEUS and B-Mode Ultrasound Images Using Advanced Machine Learning Methods.

Hepatocellular Carcinoma (HCC) is the most common malignant liver tumor, being present in 70% of liver cancer cases. It usually evolves on the top of the cirrhotic parenchyma. The most reliable method for HCC diagnosis is the needle biopsy, which is an invasive, dangerous method. In our research, specific techniques for non-invasive, computerized HCC diagnosis are developed, by exploiting the information from ultrasound images. In this work, the possibility of performing the automatic diagnosis of HCC within B-mode ultrasound and Contrast-Enhanced Ultrasound (CEUS) images, using advanced machine learning methods based on Convolutional Neural Networks (CNN), was assessed. The recognition performance was evaluated separately on B-mode ultrasound images and on CEUS images, respectively, as well as on combined B-mode ultrasound and CEUS images. For this purpose, we considered the possibility of combining the input images directly, performing feature level fusion, then providing the resulted data at the entrances of representative CNN classifiers. In addition, several multimodal combined classifiers were experimented, resulted by the fusion, at classifier, respectively, at the decision levels of two different branches based on the same CNN architecture, as well as on different CNN architectures. Various combination methods, and also the dimensionality reduction method of Kernel Principal Component Analysis (KPCA), were involved in this process. These results were compared with those obtained on the same dataset, when employing advanced texture analysis techniques in conjunction with conventional classification methods and also with equivalent state-of-the-art approaches. An accuracy above 97% was achieved when our new methodology was applied.

Integration of Real-Time Image Fusion in the Robotic-Assisted Treatment of Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide, with its mortality rate correlated with the tumor staging; i.e., early detection and treatment are important factors for the survival rate of patients. This paper presents the development of a novel visualization and detection system for HCC, which is a composing module of a robotic system for the targeted treatment of HCC. The system has two modules, one for the tumor visualization that uses image fusion (IF) between computerized tomography (CT) obtained preoperatively and real-time ultrasound (US), and the second module for HCC automatic detection from CT images. Convolutional neural networks (CNN) are used for the tumor segmentation which were trained using 152 contrast-enhanced CT images. Probabilistic maps are shown as well as 3D representation of HCC within the liver tissue. The development of the visualization and detection system represents a milestone in testing the feasibility of a novel robotic system in the targeted treatment of HCC. Further optimizations are planned for the tumor visualization and detection system with the aim of introducing more relevant functions and increase its accuracy.

Comparison of Deep-Learning and Conventional Machine-Learning Methods for the Automatic Recognition of the Hepatocellular Carcinoma Areas from Ultrasound Images.

The emergence of deep-learning methods in different computer vision tasks has proved to offer increased detection, recognition or segmentation accuracy when large annotated image datasets are available. In the case of medical image processing and computer-aided diagnosis within ultrasound images, where the amount of available annotated data is smaller, a natural question arises: are deep-learning methods better than conventional machine-learning methods? How do the conventional machine-learning methods behave in comparison with deep-learning methods on the same dataset? Based on the study of various deep-learning architectures, a lightweight multi-resolution Convolutional Neural Network (CNN) architecture is proposed. It is suitable for differentiating, within ultrasound images, between the Hepatocellular Carcinoma (HCC), respectively the cirrhotic parenchyma (PAR) on which HCC had evolved. The proposed deep-learning model is compared with other CNN architectures that have been adapted by transfer learning for the ultrasound binary classification task, but also with conventional machine-learning (ML) solutions trained on textural features. The achieved results show that the deep-learning approach overcomes classical machine-learning solutions, by providing a higher classification performance.

Concept Mapping, an Effective Tool for Long-Term Memorization of Anatomy-A Quasi-Experimental Research Carried out among 1st Year General Medicine Students.

This study is part of a doctoral thesis conducted at the Faculty of Psychology of Babes-Bolyai University in collaboration with the University of Medicine, both from Cluj-Napoca, Romania. The starting point of the study was based on the eternal question of the medical student-"How should I learn to manage to retain so much information?" This is how learning through conceptual maps and learning by understanding has been achieved. In the study, a number of 505 students from the Faculty of General Medicine were randomly selected and divided into groups, to observe changes in the grades they obtained when learning anatomy with the concept mapping method vs. traditional methods. Six months later, a retest was carried out to test long-term memory. The results were always in favor of the experimental group and were statistically significant (with one exception), most notably for the 6-month retesting. It was also observed that the language of teaching, different or the same as the first language, explains that exception, at least partially. Other results were taken into account, such as the distribution of bad and good grades in the two groups. Other parameters that influenced the obtained results and which explain some contradictory results in the literature are discussed. In conclusion, the use of conceptual maps is useful for most students, both for short and long-term memory.

Periodontal evaluation using a non-invasive imaging method (ultrasonography).

The periodontal disease and gingival bleeding are highly prevalent in the adult population worldwide. The World Health Organization (WHO) data shows that 90-100% of the 34-year-old adults present gingival inflammation. Therefore, an investigation method is required to allow the assessment of the periodontal disease as well as the monitoring of the evolution of the gingival inflammation after periodontal treatments. Non-invasive and operator-independent methods for periodontal examination are necessary for diagnosing and monitoring the periodontal disease. The periodontal ultrasonography is a reliable technique for visualizing the anatomical elements which are necessary to diagnose the periodontal status. Using this imaging technique the dentino-enamel junction, the cortical bone, the radicular surface from the crown to the alveolar bone, the gingival tissue can be seen without interfering with those elements during the examination. Also, calculus visualization is possible before and after scaling in order to evaluate the quality of the treatment. Using 2D ultrasonography is not feasible in dental practice as it requires extensive experience and is also time consuming. The reproducibility of the 2D slices is very difficult in order to have the possibility to compare different investigations efficiently. 3D reconstructions of the periodontal tissue can be a very good alternative to eliminate the operator dependence. Ultrasonography allows the practitioner to visualize the anatomic elements involved in making a periodontal diagnosis. It also allows tracking of subsequent changes. This method is not commonly used for periodontal examination and further studies are required. Previous studies show that ultrasonography can be a reliable non-invasive method to diagnose and monitor the periodontal disease.

The potential of ultrasonography in the evaluation of foot orthotics therapy.

Foot orthotics prescription is based on the foot functioning paradigms with tissue stress theory being in avant-garde. The main goal of orthotic therapy is to reduce the internal tissue's pathological stresses in the foot structures. Traditionally, ultrasound scanning technique depicts anatomic related data of both common and uncommon pathology encountered in the clinical practice, helping in diagnosing, treating and evaluating, which are equally important for the practitioners. Its accessibility, compared to biomechanical modelling, makes this technique a valuable tool in the field of foot and ankle disorders. Despite its user-dependent limitation, the ongoing technical progress improves the ability of ultrasonography as a highly advanced procedure in musculoskeletal imaging, being also a valuable searching tool for musculotendinous mechanics or morphological changes as a result of a conservative intervention. The aim of the present work was to perform a review of the state of the art concerning the usefulness of ultrasonography in the study of foot orthotic therapy and to analyze its effectiveness.

Computer-assisted identification of the gingival sulcus and periodontal epithelial junction on high-frequency ultrasound images.

UNLABELLED: The primary aim of this study was to demonstrate that periodontal ultrasonography is a reliable method with which to identify and evaluate the attachment level of the gingival junctional epithelium. A secondary aim was to devise an automated computer-assisted method that allows the examiner to more easily identify the gingival sulcus contour on ultrasound images. MATERIAL AND METHODS: This in vitro study was carried out on 36 sites on the lingual surface of eight pig mandibles. For each site, periodontal ultrasonography was performed by the same examiner, using DermaScan C Cortex Technology (Denmark) with a 20-MHz transducer. Subsequently, the mandibles were sectioned with a microtome and examined by direct microscopy. To facilitate identification of the gingival sulcus on ultrasound images, a computational method was adopted. RESULTS: Computer processing of the ultrasound images slightly modified the contour of the gingival sulcus. The absolute mean differences in the linear measurements of the Dermascan-automated computer-generated values and the corresponding values of microscopy, which is the gold standard, varied between 0.06 and 1.75 mm. Statistical analysis showed that with respect to the gingival sulcus height, the correlation between the computer-processed ultrasound images and the direct microscopy images was stronger than the correlation between the non-processed ultrasound images and those from direct microscopy. CONCLUSIONS: Ultrasonographic examination of the periodontal tissues allows the examiner to localize the gingival epithelial attachment level and provides substantial data regarding the soft gingival tissues.

Iterative methods for obtaining energy-minimizing parametric snakes with applications to medical imaging.

After a brief survey on the parametric deformable models, we develop an iterative method based on the finite difference schemes in order to obtain energy-minimizing snakes. We estimate the approximation error, the residue, and the truncature error related to the corresponding algorithm, then we discuss its convergence, consistency, and stability. Some aspects regarding the prosthetic sugical methods that implement the above numerical methods are also pointed out.

Abdominal tumor characterization and recognition using superior-order cooccurrence matrices, based on ultrasound images.

The noninvasive diagnosis of the malignant tumors is an important issue in research nowadays. Our purpose is to elaborate computerized, texture-based methods for performing computer-aided characterization and automatic diagnosis of these tumors, using only the information from ultrasound images. In this paper, we considered some of the most frequent abdominal malignant tumors: the hepatocellular carcinoma and the colonic tumors. We compared these structures with the benign tumors and with other visually similar diseases. Besides the textural features that proved in our previous research to be useful in the characterization and recognition of the malignant tumors, we improved our method by using the grey level cooccurrence matrix and the edge orientation cooccurrence matrix of superior order. As resulted from our experiments, the new textural features increased the malignant tumor classification performance, also revealing visual and physical properties of these structures that emphasized the complex, chaotic structure of the corresponding tissue.

Ultrasonography contribution to the detection and characterization of hepatic restructuring: is the "virtual biopsy" taken into consideration?

Generally, the evolution of diffuse liver diseases is variable but quite long. Even the severe types of chronic hepatitis have a slow progression which implies decades, often over 20-30 years. Cirrhosis is the principal long time complication of chronic hepatopathies. It represents a major risk factor for the development of hepatocellular carcinoma. Ultrasonography plays an important role among the methods used for detecting diffuse liver diseases, for placing them and identifying supplementary risk factors for carcinogenesis and of hepatocellular carcinoma itself. The two- and especially the three-dimensional exploration allow the characterization of hepatic texture and the identification of certain changes which may suggest hepatic restructuring.