Clinical, dermoscopic and histological assessment of melanocytic lesions: a comparative study of the accuracy of the diagnostic methods.

BACKGROUND:   Worldwide, the incidence of melanoma is increasing, while late diagnosis is related to poor prognosis. A significant risk marker for melanoma is the presence of atypical nevi; therefore, it is of outstanding importance to make accurate clinical classification of common benign nevi, atypical nevi, and melanomas. The non-invasive method of dermoscopy allowed for the visualization of structures invisible to the naked eye and undoubtedly advanced the assessment of melanocytic lesions to a new dimension. This study aimed to evaluate the sensitivity and specificity of naked-eye examination and dermoscopy in diagnosing melanocytic lesions compared to the histopathological results, constituting the gold standard of diagnosis. MATERIAL AND METHODS: One hundred eighteen melanocytic lesions were clinically evaluated via the naked eye and dermoscopic examination, using Pattern Analysis Methodology, and afterward, they were excised. The histopathological results were correlated with the findings. RESULTS: According to the final histopathological analysis, 63 common benign nevi, 41 dysplastic nevi, and 14 cutaneous melanomas were excised in total. Clinical examination via the naked eye showed 78.2 % sensitivity and 71.4 % specificity in identifying the clinical atypia, while dermoscopy demonstrated 89.1 % sensitivity and 93.7 % specificity. CONCLUSIONS: The results of the present study indicate a higher sensitivity and specificity of dermoscopy in evaluating and diagnosing melanocytic lesions compared to the naked-eye examination. HIPPOKRATIA 2021, 25 (4):156-161.

Assessing the performance of four different categories of histological criteria in brain tumours grading by means of a computer-aided diagnosis image analysis system.

Brain tumours are considered one of the most lethal and difficult to treat forms of cancer, with unknown aetiology and lack of any realistic screening. In this study, we examine, whether the combination of descriptive criteria, used by expert histopathologists in assessing histologic tissue samples, and quantitative image analysis features may improve the diagnostic accuracy of brain tumour grading. Data comprised 61 cases of brain cancers (astrocytomas, oligodendrogliomas, meningiomas) collected from the archives of the University Hospital of Patras, Greece. Incorporating physician's descriptive criteria and image analysis's quantitative features into a discriminant function, a computer-aided diagnosis system was designed for discriminating low-grade from high-grade brain tumours. Physician's descriptive features, when solely used in the system, proved of high discrimination accuracy (93.4%). When verbal descriptive features were combined with quantitative image analysis features in the system, discrimination accuracy improved to 98.4%. The generalization of the proposed system to unseen data converged to an overall prediction accuracy of 86.7% ± 5.4%. Considering that histological grading affects treatment selection and diagnostic errors may be notable in clinical practice, the utilization of the proposed system may safeguard against diagnostic misinterpretations in every day clinical practice.

Breast cancer characterization based on image classification of tissue sections visualized under low magnification.

Rapid assessment of tissue biopsies is a critical issue in modern histopathology. For breast cancer diagnosis, the shape of the nuclei and the architectural pattern of the tissue are evaluated under high and low magnifications, respectively. In this study, we focus on the development of a pattern classification system for the assessment of breast cancer images captured under low magnification (×10). Sixty-five regions of interest were selected from 60 images of breast cancer tissue sections. Texture analysis provided 30 textural features per image. Three different pattern recognition algorithms were employed (kNN, SVM, and PNN) for classifying the images into three malignancy grades: I-III. The classifiers were validated with leave-one-out (training) and cross-validation (testing) modes. The average discrimination efficiency of the kNN, SVM, and PNN classifiers in the training mode was close to 97%, 95%, and 97%, respectively, whereas in the test mode, the average classification accuracy achieved was 86%, 85%, and 90%, respectively. Assessment of breast cancer tissue sections could be applied in complex large-scale images using textural features and pattern classifiers. The proposed technique provides several benefits, such as speed of analysis and automation, and could potentially replace the laborious task of visual examination.

Preliminary images from an adaptive imaging system.

I-ImaS (Intelligent Imaging Sensors) is a European project aiming to produce real-time adaptive X-ray imaging systems using Monolithic Active Pixel Sensors (MAPS) to create images with maximum diagnostic information within given dose constraints. Initial systems concentrate on mammography and cephalography. In our system, the exposure in each image region is optimised and the beam intensity is a function of tissue thickness and attenuation, and also of local physical and statistical parameters in the image. Using a linear array of detectors, the system will perform on-line analysis of the image during the scan, followed by optimisation of the X-ray intensity to obtain the maximum diagnostic information from the region of interest while minimising exposure of diagnostically less important regions. This paper presents preliminary images obtained with a small area CMOS detector developed for this application. Wedge systems were used to modulate the beam intensity during breast and dental imaging using suitable X-ray spectra. The sensitive imaging area of the sensor is 512 x 32 pixels 32 x 32 microm(2) in size. The sensors' X-ray sensitivity was increased by coupling to a structured CsI(Tl) scintillator. In order to develop the I-ImaS prototype, the on-line data analysis and data acquisition control are based on custom-developed electronics using multiple FPGAs. Images of both breast tissues and jaw samples were acquired and different exposure optimisation algorithms applied. Results are very promising since the average dose has been reduced to around 60% of the dose delivered by conventional imaging systems without decrease in the visibility of details.

A morphological index for assessing hip osteoarthritis severity from radiographic images.

A new method is proposed for assessing the severity of hip osteoarthritis (OA) based on radiographic hip joint space (HJS) morphology. 64 hips of patients with verified unilateral OA or bilateral OA were studied by digitizing the corresponding pelvic radiographs. Radiographic OA severity was assessed employing the Kellgren and Lawrence (KL) scale. Using custom-developed software, radiographs were enhanced, the margins of both HJSs were outlined, and 64 regions of interest (ROIs), corresponding to the delineated HJSs, were obtained. Employing custom-developed algorithms, an index ("joint space morphological index" - JSMI) evaluating alterations in the shape and size of HJS was introduced, calculated and normalized with respect to each patient's individual anatomy. The JSMI values were used to introduce classification rules concerning the characterization of a hip in accordance with the KL scale. For each patient in the unilateral OA group, the OA severity was expressed as the percentage of the HJS area difference between the patient's osteoarthritic and contralateral normal hip. The per cent HJS area difference and the JSMI values were used in the design of a regression model for providing a quantitative estimation of OA severity. The per cent HJS area difference correlated highly with the pathological JSMI values (r = -0.83, p<0.001). The implementation of the JSMI-based classification rules resulted in high classification accuracies for characterizing hips as normal or osteoarthritic, 90.6% (95% exact confidence interval (CI): 80.7-96.5%), as well as for discriminating among OA severity categories, 91.7% (95% CI: 77.5-98.2%). Additionally, a simplified approach of JSMI calculation is suggested for daily clinical use. These JSMI values (JSMI simplified) were found not to differ significantly from (p>0.05), and to be strongly correlated with (r = 0.96, p<0.001), the corresponding ones obtained by the computerized approach. Additionally, the implementation of classification rules based on JSMI simplified resulted in classification accuracies identical to the corresponding ones obtained for the JSMI-based rules. The proposed method may be utilized for evaluating OA and monitoring OA progression.

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

A wavelet-based method for speckle suppression in ultrasound images of the thyroid gland is introduced. The classification of image pixels as speckle or part of important image structures is accomplished within the framework of back-propagation tracking and singularity detection of wavelet transform modulus maxima, derived from inter-scale analysis. A comparative study with other de-speckling techniques, employing quantitative indices, demonstrated that our method achieved superior speckle reduction performance and edge preservation properties. Moreover, a questionnaire regarding qualitative imaging parameters, emanating from various visual observations, was employed by two experienced physicians in order to evaluate the algorithm's speckle suppression efficiency.

Assessing hip osteoarthritis severity utilizing a probabilistic neural network based classification scheme.

A computer-based classification system is proposed for the characterization of hips from pelvic radiographs as normal or osteoarthritic and for the discrimination among various grades of osteoarthritis (OA) severity. Pelvic radiographs of 18 patients with verified unilateral hip OA were evaluated by three experienced physicians, who assessed OA severity employing the Kellgren and Lawrence scale as: normal, mild/moderate and severe. Five run-length, 75 Laws' and 5 novel textural features were extracted from the digitized radiographic images of each patient's osteoarthritic and contralateral normal hip joint spaces (HJSs). Each one of the three sets of textural features (run-lengths, Laws' and novel features) was separately utilized for assigning hips into the three OA severity categories, by means of a probabilistic neural network (PNN) classifier based hierarchical tree structure. The highest classification accuracy (100%) for characterizing hips as normal, of mild/moderate or of severe OA was obtained for the novel textural features set. Additionally, the novel textural features were used to design a mathematical regression model for providing a quantitative estimation of OA severity. Measured OA severity values, as expressed by HJS-narrowing, correlated highly (r=0.85, p<0.001) with the predicted values by the mathematical regression model. The proposed system may be valuable in OA-patient management.

A hybrid multi-scale model for thyroid nodule boundary detection on ultrasound images.

A hybrid model for thyroid nodule boundary detection on ultrasound images is introduced. The segmentation model combines the advantages of the "á trous" wavelet transform to detect sharp gray-level variations and the efficiency of the Hough transform to discriminate the region of interest within an environment with excessive structural noise. The proposed method comprise three major steps: a wavelet edge detection procedure for speckle reduction and edge map estimation, based on local maxima representation. Subsequently, a multiscale structure model is utilised in order to acquire a contour representation by means of local maxima chaining with similar attributes to form significant structures. Finally, the Hough transform is employed with 'a priori' knowledge related to the nodule's shape in order to distinguish the nodule's contour from adjacent structures. The comparative study between our automatic method and manual delineations demonstrated that the boundaries extracted by the hybrid model are closely correlated with that of the physicians. The proposed hybrid method can be of value to thyroid nodules' shape-based classification and as an educational tool for inexperienced radiologists.

Osteoarthritis severity of the hip by computer-aided grading of radiographic images.

A computer-aided classification system was developed for the assessment of the severity of hip osteoarthritis (OA). Sixty-four radiographic images of normal and osteoarthritic hips were digitized and enhanced. Employing the Kellgren and Lawrence scale, the hips were grouped by three experienced orthopaedists into three OA-severity categories: Normal, Mild/Moderate and Severe. Utilizing custom-developed software, 64 ROIs corresponding to the radiographic Hip Joint Spaces were manually segmented and novel textural features were generated. These features were used in the design of a two-level classification scheme for characterizing hips as normal or osteoarthritic (1st level) and as of Mild/Moderate or Severe OA (2nd level). At each classification level, an ensemble of three classifiers was implemented. The proposed classification scheme discriminated correctly all normal hips from osteoarthritic hips (100% accuracy), while the discrimination accuracy between Mild/Moderate and Severe osteoarthritic hips was 95.7%. The proposed system could be used as a diagnosis decision-supporting tool.

Quantitative assessment of hip osteoarthritis based on image texture analysis.

A non-invasive method was developed to investigate the potential capacity of digital image texture analysis in evaluating the severity of hip osteoarthritis (OA) and in monitoring its progression. 19 textural features evaluating patterns of pixel intensity fluctuations were extracted from 64 images of radiographic hip joint spaces (HJS), corresponding to 32 patients with verified unilateral or bilateral OA. Images were enhanced employing custom developed software for the delineation of the articular margins on digitized pelvic radiographs. The severity of OA for each patient was assessed by expert orthopaedists employing the Kellgren and Lawrence (KL) scale. Additionally, an index expressing HJS-narrowing was computed considering patients from the unilateral OA-group. A textural feature that quantified pixel distribution non-uniformity (grey level non-uniformity, GLNU) demonstrated the strongest correlation with the HJS-narrowing index among all extracted features and utilized in further analysis. Classification rules employing GLNU feature were introduced to characterize a hip as normal or osteoarthritic and to assign it to one of three severity categories, formed in accordance with the KL scale. Application of the proposed rules resulted in relatively high classification accuracies in characterizing a hip as normal or osteoarthritic (90.6%) and in assigning it to the correct KL scale category (88.9%). Furthermore, the strong correlation between the HJS-narrowing index and the pathological GLNU (r = -0.9, p<0.001) was utilized to provide percentages quantifying hip OA-severity. Texture analysis may contribute in the quantitative assessment of OA-severity, in the monitoring of OA-progression and in the evaluation of a chondroprotective therapy.

A theoretical model evaluating the angular distribution of luminescence emission in X-ray scintillating screens.

The aim of this study was to examine the angular distribution of the light emitted from radiation-excited scintillators in medical imaging detectors. This distribution diverges from Lambert's cosine law and affects the light emission efficiency of scintillators, hence it also affects the dose burden to the patient. In the present study, the angular distribution was theoretically modeled and was used to fit experimental data on various scintillator materials. Results of calculations revealed that the angular distribution is more directional than that predicted by Lambert's law. Divergence from this law is more pronounced for high values of light attenuation coefficient and thick scintillator layers (screens). This type of divergence reduces light emission efficiency and hence it increases the incident X-ray flux required for a given level of image brightness.

Identifying Differences in the P600 Component of ERP-Signals between OCD Patients and Controls Employing a PNN-based Majority Vote Classification Scheme.

In the present study an attempt was made to focus in the differences between Obsessive-Compulsive Disorder (OCD) patients and healthy controls, as reflected by the P600 component of event-related potential (ERP) signals, to locate brain areas that may be related to Working Memory (WM) deficits. Neuropsychological research has yielded contradicting results regarding WM in OCD. Eighteen patients with OCD symptomatology and 20 normal controls (age and sex matched) were subjected to a computerized version of the digit span Wechsler test. EEG activity was recorded from 15 scalp electrodes (leads). A dedicated computer software was developed to read the ERP signals and to calculate features related to the ERP P600 component (500-800 ms). Nineteen features were generated, from each ERP-signal and each lead, and were employed in the design of the Probabilistic Neural Network (PNN) classifier. Highest single-lead precision (86.8%) was found at the Fp2 and C6 leads. When the output from all single-lead PNN classifiers fed a Majority Vote Engine (MVE), the system classified correctly all subjects, providing a powerful classification scheme. Findings indicated that OCD patients differed from normal controls at the prefrontal and temporo-central brain regions.

Upgrading undergraduate biomedical engineering laboratory training.

This paper presents the upgrading of biomedical engineering laboratory training at the Department of Medical Instrumentation Technology of the Technological Educational Institution of Athens (TEI-A), taking place in the framework of the "Upgrading of Undergraduate Curricula of TEI-A" project. The educational material of selected specialized laboratory sectors is totally renewed, and new sectors are introduced, so that student-centered learning is promoted utilizing advanced computer-enhanced educational environments. The current implementation status is presented for the laboratories dealing with biosignal acquisition, medical data digital processing and, more extensively, computer networks applications in medicine, where a training application simulating a Radiology Department computer network was developed. Benefits of the use of a balanced training approach, combining hands-on experience with computer simulations, are discussed.

An image-analysis system based on support vector machines for automatic grade diagnosis of brain-tumour astrocytomas in clinical routine.

An image-analysis system based on the concept of Support Vector Machines (SVM) was developed to assist in grade diagnosis of brain tumour astrocytomas in clinical routine. One hundred and forty biopsies of astrocytomas were characterized according to the WHO system as grade II, III and IV. Images from biopsies were digitized, and cell nuclei regions were automatically detected by encoding texture variations in a set of wavelet, autocorrelation and parzen estimated descriptors and using an unsupervised SVM clustering methodology. Based on morphological and textural nuclear features, a decision-tree classification scheme distinguished between different grades of tumours employing an SVM classifier. The system was validated for clinical material collected from two different hospitals. On average, the SVM clustering algorithm correctly identified and accurately delineated 95% of all nuclei. Low-grade tumours were distinguished from high-grade tumours with an accuracy of 90.2% and grade III from grade IV with an accuracy of 88.3% The system was tested in a new clinical data set, and the classification rates were 87.5 and 83.8%, respectively. Segmentation and classification results are very encouraging, considering that the method was developed based on every-day clinical standards. The proposed methodology might be used in parallel with conventional grading to support the regular diagnostic procedure and reduce subjectivity in astrocytomas grading.

Development of the cubic least squares mapping linear-kernel support vector machine classifier for improving the characterization of breast lesions on ultrasound.

An efficient classification algorithm is proposed for characterizing breast lesions. The algorithm is based on the cubic least squares mapping and the linear-kernel support vector machine (SVM(LSM)) classifier. Ultrasound images of 154 confirmed lesions (59 benign and 52 malignant solid masses, 7 simple cysts, and 32 complicated cysts) were manually segmented by a physician using a custom developed software. Texture and outline features and the SVM(LSM) algorithm were used to design a hierarchical tree classification system. Classification accuracy was 98.7%, misdiagnosing 1 malignant an 1 benign solid lesions only. This system may be used as a second opinion tool to the radiologists.

Design and implementation of an SVM-based computer classification system for discriminating depressive patients from healthy controls using the P600 component of ERP signals.

A computer-based classification system has been designed capable of distinguishing patients with depression from normal controls by event-related potential (ERP) signals using the P600 component. Clinical material comprised 25 patients with depression and an equal number of gender and aged-matched healthy controls. All subjects were evaluated by a computerized version of the digit span Wechsler test. EEG activity was recorded and digitized from 15 scalp electrodes (leads). Seventeen features related to the shape of the waveform were generated and were employed in the design of an optimum support vector machine (SVM) classifier at each lead. The outcomes of those SVM classifiers were selected by a majority-vote engine (MVE), which assigned each subject to either the normal or depressive classes. MVE classification accuracy was 94% when using all leads and 92% or 82% when using only the right or left scalp leads, respectively. These findings support the hypothesis that depression is associated with dysfunction of right hemisphere mechanisms mediating the processing of information that assigns a specific response to a specific stimulus, as those mechanisms are reflected by the P600 component of ERPs. Our method may aid the further understanding of the neurophysiology underlying depression, due to its potentiality to integrate theories of depression and psychophysiology.

Using handheld devices for real-time wireless teleconsultation.

Recent advances in the hardware of handheld devices, opened up the way for newer applications in the healthcare sector, and more specifically, in the teleconsultation field. Out of these devices, this paper focuses on the services that personal digital assistants and smartphones can provide to improve the speed, quality and ease of delivering a medical opinion from a distance and laying the ground for an all-wireless hospital. In that manner, PDAs were used to wirelessly support the viewing of digital imaging and communication in medicine (DICOM) images and to allow for mobile videoconferencing while within the hospital. Smartphones were also used to carry still images, multiframes and live video outside the hospital. Both of these applications aimed at increasing the mobility of the consultant while improving the healthcare service.

Upgrading the biomedical engineering undergraduate curriculum based on current trends in higher education.

This paper presents the reforming of the curriculum of the Department of Medical Instrumentation Technology at the Technological Educational Institution of Athens (TEI-A), as inspired by current trends in higher education. The reforming is taking place in the framework of the "Upgrading of Undergraduate Curricula of TEI-A" project The project-funded upgrading focuses on a core of eight laboratory sectors, with particular emphasis placed on student-centered learning, taking advantage of computer-enhanced educational environment. The existing and proposed curricula are compared. The student workload in the proposed curriculum is reduced, while maintaining an extensive set of basic and applied knowledge related to biomedical engineering. The overall aim is to provide a curriculum that will help in developing multi-skilled individuals that can relate to the demands of this field within a dynamic social and economical environment.

Support vector machines based analysis of brain SPECT images for determining cerebral abnormalities in asymptomatic diabetic patients.

PURPOSE: An image processing method was developed to investigate whether brain SPECT images of patients with diabetes mellitus type II (DMII) and no brain damage differ from those of normal subjects. MATERIALS AND METHODS: Twenty-five DMII patients and eight healthy volunteers underwent brain 99mTc-Bicisate SPECT examination. A semi-automatic method, allowing for physician's interaction, was developed to delineate specific brain regions (ROIs) on the SPECT images. Twenty-eight features from the grey-level histogram and the spatial-dependence matrix were computed from numerous small image-samples collected from each specific ROI. Classification into 'diabetics' and 'non-diabetics' was performed for each ROI separately. The classical least squares-minimum distance (LSMD) classifier and the recently developed support vector machines (SVM) classifier were used. System performance was evaluated by means of the leave-one-out method; one sample was left out, the classifier was trained by the rest of the samples, and the left-out sample was classified. By repeating for all samples, the classifier's performance could be tested on data not incorporated in its design. RESULTS: Highest classification accuracies (LSMD: 97.8%, SVM: 99.1%) were achieved at the right occipital lobule employing two features, the standard deviation and entropy. For the rest of the ROIs classification accuracies ranged between 84.5 and 98.6%. CONCLUSION: Our findings indicate cerebral blood flow disruption in patients with DMII. The proposed system may assist physicians in evaluating cerebral blood flow in patients with DMII undergoing brain SPECT.

Integrated model for estimating phosphor signal and noise transfer characteristics on medical images: application to CdPO3Cl:Mn phosphor screens.

An integrated model describing the signal and noise transfer characteristics of the objective image quality and information content in phosphor-produced images is presented. In the context of this model, important imaging parameters, namely optical gain, modulation transfer function, noise transfer function, detective quantum efficiency and information capacity were experimentally evaluated using seven laboratory-prepared CdPO3Cl:Mn test phosphor screens of varying coating thickness. This phosphor has been previously shown to exhibit high spectral compatibility properties with the films and optical sensors used in digital imaging systems. Experiments were performed using 50-120 kVp X-rays produced by a medical X-ray unit. Results showed that, for thick screens, optical gain attained peak values close to 200 optical photons per incident X-ray at 50 kVp. The noise transfer function was higher than the modulation transfer function. For the thin screen of 21 mg cm-2, the modulation transfer function was 0.25 at 100 line pairs mm-1, and the corresponding noise transfer function was 0.4. The detection quantum efficiency peak value was 0.22 at 50 kVp. These values are within acceptable performance limits, and, given the phosphor material's high spectral compatibility and medium temporal response, CdPO3Cl:Mn could be considered for use in X-ray detectors of static radiography imaging.