Wyniki wyszukiwania - BazTech

1

Ensemble of classifiers based on deep learning for medical image recognition

Gil Fabian, Osowski Stanisław, Świderski Bartosz, Słowińska Monika

Metrology and Measurement Systems

|

2023

|

Vol. 30, nr 1

139--156

EN

The paper presents special forms of an ensemble of classifiers for analysis of medical images based on application of deep learning. The study analyzes different structures of convolutional neural networks applied in the recognition of two types of medical images: dermoscopic images for melanoma and mammograms for breast cancer. Two approaches to ensemble creation are proposed. In the first approach, the images are processed by a convolutional neural network and the flattened vector of image descriptors is subjected to feature selection by applying different selection methods. As a result, different sets of a limited number of diagnostic features are generated. In the next stage, these sets of features represent input attributes for the classical classifiers: support vector machine, a random forest of decision trees, and softmax. By combining different selection methods with these classifiers an ensemble classification system is created and integrated by majority voting. In the second approach, different structures of convolutional neural networks are directly applied as the members of the ensemble. The efficiency of the proposed classification systems is investigated and compared to medical data representing dermoscopic images of melanoma and breast cancer mammogram images. Thanks to fusion of the results of many classifiers forming an ensemble, accuracy and all other quality measures have been significantly increased for both types of medical images.

2

Neural networks from Keras in skin lesion diagnostic

Michalska-Ciekańska Magdalena

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2022

|

T. 12, nr 1

40--43

EN

Melanoma is currently one of the most dangerous skin diseases, in addition many others appear in the population. Scientists are developing techniques for early non-invasive skin lesions diagnosis from dermatoscopic images, for this purpose neural networks are increasingly used. Many tools are being developed to allow for faster implementation of the network, including the Keras package. The article presents selected methods of diagnosing skin diseases, including the process of classification, features selection, extracting the skin lesion from the whole image.The described methods have been implemented using deep neural networks available in the Keras package. The article draws attention to the effectiveness, specificity, accuracy of classification based on available data sets, attention was paid to tools that allow for more effective operation of algorithms.

PL

Melanoma jest obecnie jedną z najbardziej niebezpiecznych chorób skóry, oprócz niej pojawia się w populacji wiele innych. Naukowcy rozwijają techniki wczesnego nieinwazyjnego diagnozowania zmian skórnych z obrazów dermatoskopowych, w tym celu coraz częściej wykorzystywane są sieci neuronowe. Powstaje wiele narzędzi powzalajcych na szybszą implementację sieci należy do niej pakiet Keras. W artykule przedstawiono wybrane metody diagnostyki chorób skóry, należy do nich proces klasyfikacji, selekcji cech, wyodrębnienia zmiany skórnej z całego obrazu. Opisane metody zostały zostały zaimplementowane za pomocą dostępnych w pakiecie Keras głębokich sieci neuronowych. W artykule zwrócono uwagę na skuteczność, specyficzność, dokładność klasyfikacji w oparciu o dostępne zestawy danych, zwrócono uwagę na narzędzi pozwalające na efektywniejsze działanie algorytmów.

3

Transcriptomic data analysis of melanocytes and melanoma cell lines of LAT transporter genes for precise medicine

Szczepanek Monika, Panek Dominik, Przybyło M., Moskal Paweł, Stępień Ewa Ł.

Bio-Algorithms and Med-Systems

|

2022

|

Vol. 18, no. 1

144--150

EN

Background: Boron Neutron Capture Therapy (BNCT) is a two-step treatment that can be used in some types of cancers. It involves administering a compound containing boron atoms to the patient and irradiating the affected area of the body with a neutron beam. The success of the therapy depends mainly on the delivery of the boron isotope (10B) to the tumor using an appropriate boron carrier. One of the boron carriers used is boronophenylalanine (BPA). Therefore, in research on the use of boron carriers, it is also important to know the mechanisms of its uptake by cells. Aim: To study the expression of LAT family genes in two melanoma (high melanotic WM115 and low melanotic WM266-4) cell lines and melanocytes (HEMa-Lp) which are responsible for the transport the BPA into cells. Methods: To normalize data from the transcriptomic analysis, the ratio of the median method was used. This allowed the samples to be compared with each other. Comparison metrics included log-fold change (LFC) values. The heatmap of LFC values and the cluster map were created. These graphs show the similarities and differences between the samples. Results: Transcriptomic data show that in melanocytes, LFC for SLC7A5 (LAT1) and SLC3A2 (4Fhc) was higher than in melanoma cell lines, which corresponded with their melanin content. Conclusion: Our results indicate overexpression of BPA transporter genes in normal cells (melanocytes), which may suggest the highest level of these proteins in melanocytes compared to less melanotic melanoma. Therefore, for BNCT, the use of BPA as the 10B carrier will require additional qualifying tests of amino acid transporter expression for patients and specific tumors to develop a personalized BNCT.

4

Selected applications of deepneural networks in skin lesion diagnostic

Michalska-Ciekańska Magdalena

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2021

|

T. 11, nr 4

18--21

EN

The article provides an overview of selected applications of deep neural networks in the diagnosis of skin lesions from human dermatoscopic images, including many dermatological diseases, including very dangerous malignant melanoma. The lesion segmentation process, features selectionand classification was described.Application examples of binary and multiclass classification are given.The described algorithms have been widely used in the diagnosis of skin lesions. The effectiveness, specificity, and accuracy of classifiers were compared and analyzed based on available datasets.

PL

Artykuł zawiera przeglądwybranychzastosowań głębokich sieci neuronowych w diagnostyce zmian skórnych zobrazów dermatoskopowych człowieka z uwzględnieniem wielu choróbdermatologicznych, w tym bardzo niebezpiecznejz nich malignant melanoma. Został opisany processegmentacjizmiany, selekcji cech i klasyfikacji. Uwzględniono przykłady binarnej i wieloklasowej klasyfikacji. Opisane algorytmy znalazły szerokie zastosowanie w diagnostyce zmian skórnych.Porównano i przeanalizowanoskuteczność, specyficznośći dokładność klasyfikatorów w oparciu o dostępne zestawy danych.

5

Analyses of skin lesion areas after thresholding

Michalska Magdalena

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2020

|

T. 10, nr 3

9--12

EN

Melanoma is one of the fastest spreading cancers.The aim of the article is to segment the skin lesionsfrom human skin dermatoscopic images covered by melanoma. Threshold segmentation was used, which allows a single skin lesionto be analyzed. Itshows the four areas of each based on their color. The created software monitors the border of skin lesion areas.Segmentation and analysis of the resulting images with different areas of skin change was carried out in the Matlab software.

PL

Czerniak to jeden z najszybciej rozprzestrzeniających się nowotworów. Celem artykułu jest segmentacja zmiany skórnej z obrazów dermatoskopowych ludzkiej skóry objętych czerniakiem. Użyto segmentacj przez progowanie, która pozwala na analizę pojedyńczejzmiany skórnej. Ukazuje cztery obszary każdej z nich w oparciu o ich barwę. Stworzone oprogramowanie monitoruje granicę obszarów zmiany skórnej. Segmentacjai analiza powstałych obrazów z różnymi obszarami zmiany skórnej została przeprowadzona w środowisku Matlab.

6

Monitoring vital functions of A-375 melanoma cell cultures via thin-film nickel capacitors

Wilczyńska Aleksandra, Kociubiński Andrzej, Zarzeczny Dawid, Szypulski Maciej, Pigoń Dominika, Małecka-Massalska Teresa, Prendecka-Wróbel Monika

Engineering of Biomaterials

|

2020

|

Vol. 23, no. 157

10--14

EN

This article deals in the constantly developing branch of microelectronic devices used in various fields of medicine, i.e. diagnostics and treatment of previously incurable human diseases. A method for assessing and monitoring the vital functions of living cells by measuring cellular impedance in real-time using the ECIS® system and a commercial culture substrate is presented. The goal was to develop a substrate significantly less expensive than a commercial substrate that would be suitable for multiple uses and compatible with the ECIS® measurement station. Moreover, thanks to the use of a material with electrochemical properties other than the biocompatible material (gold or platinum) it is possible to observe the cells behavior with regard to the toxic agent. For this purpose, a culture substrate with nickel comb capacitors was used. To make the electrodes, a thin metal layer was sputtered on polycarbonate plates in the magnetron sputtering process. Prior to the next stages, technological masks were designed so as to fit in the ECIS® measuring station. Subsequently, the microelectronic processes of photolithography and etching the metal layer were performed. Finally, the wells were glued onto the culture medium with a biocompatible adhesive. The completed substrates were transferred to the Department of Human Physiology, Medical University of Lublin, for the culture test on A-375 human melanoma cells. The results of the experiment determined the usefulness of the device for monitoring cell culture vital functions by means of impedance measurement.

7

Skin Lesion Analysis Toward Melanoma Detection Using Deep Learning Techniques

Sherif Fatma, Mohamed Wael A., Mohra A.S.

International Journal of Electronics and Telecommunications

|

2019

|

Vol. 65, No. 4

597--602

EN

In the last few years, a great attention was paid to the deep learning Techniques used for image analysis because of their ability to use machine learning techniques to transform input data into high level presentation. For the sake of accurate diagnosis, the medical field has a steadily growing interest in such technology especially in the diagnosis of melanoma. These deep learning networks work through making coarse segmentation, conventional filters and pooling layers. However, this segmentation of the skin lesions results in image of lower resolution than the original skin image. In this paper, we present deep learning based approaches to solve the problems in skin lesion analysis using a dermoscopic image containing skin tumor. The proposed models are trained and evaluated on standard benchmark datasets from the International Skin Imaging Collaboration (ISIC) 2018 Challenge. The proposed method achieves an accuracy of 96.67% for the validation set. The experimental tests carried out on a clinical dataset show that the classification performance using deep learning-based features performs better than the state-of-the-art techniques.

8

Improvement in the diagnosis of melanoma and dysplastic lesions by introducing ABCD-PDT features and a hybrid classifier

Zakeri A., Hokmabadi A.

Biocybernetics and Biomedical Engineering

|

2018

|

Vol. 38, no. 3

456--466

EN

Melanoma and dysplastic lesions are pigmented skin lesions whose accurate classification is of great importance. In this paper, we have proposed a computer-aided diagnosis (CAD) system to improve the diagnostic ability of the conventional ABCD (asymmetry, border irregularity, color, and diameter) analysis. We introduced features extracted by local analysis of range of intensity variations within the lesion that describe pigment distribution and texture (PDT) features. The statistical distribution of pigmentation at a specified direction and distance was analyzed through grey level co-occurrence matrix (GLCM). Some other quantitative features were also extracted by computing neighborhood grey-tone difference matrix. These were correlated with human perception of texture. A hybrid classifier was designed for classification of melanoma, dysplastic, and benign lesions. Log-linearized Gaussian mixture neural network (LLGMNN), K-nearest neighborhood (KNN), linear discriminant analysis (LDA), and support vector machine (SVM) construct the hybrid classifier. The proposed system was evaluated on a set of 792 dermoscopy images and the diagnostic accuracies of 96.8%, 97.3%, and 98.8% for melanoma, dysplastic, and benign lesions were achieved, respectively. The results indicate that PDT features are promising features which in combination with the conventional ABCD features are capable of enhancing the classification performance of the pigmented skin lesions.

9

Research on melanocitic skin lesion infobase enlargment - new facts and concepts

Chrzanowska D., Niemiec R., Hippe Z.

TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk

|

2017

|

Vol. 21, No 4

391--393

EN

The melanocytic skin lesion infobase, available at http://synthesis.melanoma.pl (also http://synteza.melanoma.pl, in Polish; referred to as INP) is currently undergoing a complete modification of the way in which (i) the internal synthesis algorithms and (ii) the classification of lesions are performed. We investigated 29 new real images of melanocytic skin lesions, focusing on how humans perform classification based on experience. In conclusion we suggest to add a new color – connected with the depth of a lesion – to the K term of Asymmetry (A),Border (B) and linear Combination of colors and structures (K) method (referred to as ABK).

10

Comparison of three In silico models of heat transfer in human body exposed to mild cooling

Buliński P., Ostrowski Z., Adamczyk W.

Studia Informatica

|

2016

|

Vol. 37, nr 3A

31--42

EN

The comparison of three numerical models of skin undergoing thermal stimulation in a human forearm is presented. Small brass compress is used to cool tissues, followed by the analysis of the skin temperature recovery process. In silico generated results are validated against in vivo measurements on 8 male adults.

PL

W pracy przedstawiono porównanie trzech modeli numerycznych tkanek przedramienia poddanych stymulacji termicznej. Chłodzenie skóry zrealizowano za pomocą mosiężnego kompresu. Analizowano proces powrotu schłodzonej skóry do warunków równowagi termicznej. Wyniki symulacji in silico porównano z pomiarami in vivo wykonanymi dla grupy 8 dorosłych mężczyznach.

11

Search for Resolution Invariant Wavelet Features of Melanoma Learned by a Limited ANN Classifier

Surówka G.

Schedae Informaticae

|

2016

|

Vol. 25

189--207

EN

This article addresses the Computer Aided Diagnosis (CAD) of melanoma pigmented skin cancer. We present back-propagated Artificial Neural Network (ANN) classifiers discriminating dermoscopic skin lesion images into two classes: malignant melanoma and dysplastic nevus. Features used for our classification experiments are derived from wavelet decomposition coefficients of the image. Our research objective is i) to select the most efficient topology of the hidden layers and the network learning algorithm for full-size and downgraded image resolutions and, ii) to search for resolution-invariant topologies and learning methods. The analyzed classifiers should be fit to work on ARM-based hand-held devices, hence we take into account only limited learning setups.

12

Validation of a numerical model of locally cooled tissues

Buliński P., Ostrowski Z., Adamczyk W., Fityka A., Nowak A. J.

Computer Assisted Methods in Engineering and Science

|

2015

|

Vol. 22, no. 4

305--314

EN

Measurements of heat transfer and temporal temperature distribution can be used as input in the diagnostic tools and methods of skin lesions, with special attention paid to malignant melanoma identification. Such approach requires mutual use of skin temperature and heat flux measurements combined with numerical simulation. A mild skin cooling process by a brass compress is considered in this paper. The temperature distribution on the skin and the heat flux between metal and tissues are measured. They are used in the course of validation study of the proposed numerical model. A numerical model of heat transfer in living tissues is described by Pennes’ bioheat equation augmented with additional models of passive thermoregulation and vasoconstriction effects. The information regarding material properties of tissues and cooling compress involved in the simulation is essential to accurately solve this problem. Therefore, the main purpose of this work is to determine the accurate material property information by means of laboratory experiments.

13

Design of a teledermatology system to support the consultation of dermoscopic cases using mobile technologies and cloud platform

Jaworek-Korjakowska J., Tadeusiewicz R.

Bio-Algorithms and Med-Systems

|

2015

|

Vol. 11, no. 1

53--58

EN

Skin cancer is the most commonly diagnosed type of cancer in humans regardless of age, gender, or race. One of the most common malignant skin cancers is melanoma, which is a dangerous proliferation of melanocytes. In the last several years, an increasing melanoma incidence has been observed worldwide, and the incidence rate is increasing faster than those of any other skin cancer. The correct identification and diagnosis of moles still creates problems to inexperienced dermatologists and family physicians. In this paper, we present a new approach to the problem of assessing difficult cases in dermatology. We propose a teledermatology system to support the consultation process between family physicians and experts in the field of dermoscopic images. The system consists of a desktop monitoring application and a special smartphone application implemented for experts. If necessary, the physician can send the dermoscopic image to two dermatologists for further examination. This cloud-based architecture provides an interesting system for a fast and efficient exchange of dermatological information. Initial results and assessment of doctors are promising and indicate that the application can be used as a decision support system for dermoscopic images.

14

Wavelet based classification of skin lesion images

Surówka G., Merkwirth C., Żabińska-Płazak E., Graca A.

Bio-Algorithms and Med-Systems

|

2006

|

Vol. 2, no. 4

43--49

EN

Visual examination of the early stages of the melanocytic skin cancer (melanoma) may often lead to a false diagnosis. Only the resection and then histologic examination of the lesion can fully detect malignant transformations of the skin. This is the reason why development of non-invasive methods for dermatological diagnosis, like dermatoscopy, is of key importance. We build a MLP-based binary classifier for discriminating melanoma from dysplastic nevus utilizing textural information contained in the skin lesion images taken in dermatoscopic examinations. Our analysis is based on the multiresolution wavelet-based decomposition of the images. Significant features of both classes are found by means of the Ridge regression models. Discriminating melanoma from dysplastic nevus with this method yields a sensitivity and specificity of 89.5% and 90%, respectively.

PL

Wizualna ocena wczesnych stanów procesu nowotworzenia skóry może prowadzić do błędnej diagnozy. Jedynie resekcja oraz histologiczna ocena może ocenić obecność procesu nowotworzenia. Stąd potrzeba nieinwazyjnej oceny w dermatologii jest potrzebą chwili. Zbudowaliśmy bazujący na MLP binarny klasyfikator dla dyskryminacji melanoma w oparciu o obrazy uzyskane dermatoskopowo. Metoda bazuje na dekompozycji obrazu. Model regresji Ridge'go został zaadaptowany dla klasyfikacji obrazu co dało specyficzność oceny rzędu 89.5% i 90%.

15

Próba ilościowej oceny koncentracji porfiryn w układach biologicznych metodą obrazowania fluorescencyjnego

Sawow A., Matuszak Z., Wasilewska-Radwanska M.

Acta Bio-Optica et Informatica Medica. Inżynieria Biomedyczna

|

2006

|

Vol. 12, nr 1

VIII-3

PL

W prezentowanej pracy przedstawiono próbę podejścia do ilościowej oceny stężenia fotouczulaczy z grupy porfiryn w komórkach nowotworowych na podstawie zdjęć ich fluorescencji . Obserwowano syntezę porfiryn przez komórki nowotworowe mysiej melanomy S91 z podawanego zewnętrznie kwasu 5-aminolewulinowego (5-ALA). Fluorescencja tak zsyntetyzowanej porfiryny była obserwowana w mikroskopie fluorescencyjnym a uzyskane zdjęcia były przedmiotem analizy. Skalibrowano zależność intensywności średniej składowej R zdjęć (model RGB) od stężenia porfiryny. Pozwoliło to na korelację koncentracji fotouczulacza i obserwowanej intensywności fluorescencji w strukturach komórkowych. Uzyskane wyniki pozwoliły na ocenę ilościową stężeń fotouczalczy w badanych próbkach oraz zbadanie intensywności syntezy związku przez badane komórki. Zaproponowana metodyka może być przeniesiona na guzy nowotworowe i być pomocna w diagnostyce i praktyce klinicznej stosującej terapie fotodynamiczną.

EN

The aim of this study was to determine the quantity of the protoporphyrin IX (PpIX) in S91 melanoma cells after topical administration of its precursor 5-aminolevulinic acid (5-Ala) using the photographs of PpIX fluorescence. The dependence of mean R component of the colour cell photography ( R componenet in RGB model) on porphyrin concentration was determined and used to estimate the porphyrin concentration in S91 cells on the basis porphyrin fluorescence intensity maps. The results indicate that using this method is possible to determine the dependence of quantity of the synthesized PpIX by S91 cells on the time of incubation with 5-Ala. The proposed method can be easily modified and used in determining of photosenstizers concentration in tumours.

16

Zastosowanie videodermatoskopii w diagnostyce zmian barwnikowych skóry

Żabińska-Płazak E., Wojas-Pelc A., Dyduch G.

Bio-Algorithms and Med-Systems

|

2005

|

Vol. 1, no. 1/2

333--338

PL

Czerniak skóry należy do najbardziej złośliwych nowotworów człowieka. Obserwowany jest również na całym świecie bardzo dynamicznie narastający wzrost zachorowalności na czerniaka. Wczesna diagnostyka zmian jest warunkiem zastosowania skutecznego leczenia. Videodermatoskopia (mikroskopia epiluminescencyjna) jest metodą bardzo pomocną we wczesnej diagnostyce czerniaka skóry oraz monitorowaniu znamion atypowych. Jej dalszy rozwój jest ściśle uzależniony od rozwoju systemów komputerowych umożliwiających precyzyjną ocenę obrazów uzyskanych w badaniu dermatoskopowym.

EN

Melanoma is one of the most malignant human neoplasmas. The dynamic increase of incidence of melanoma is observed all over the world. The early diagnose of melanoma is the condition of succesful treatment. Videodermatoscopy (epiluminescence microscopy) is a very useful method in early diagnose of melanoma and monitoring of atypical nevi. The development of videodermatoscopy is strictly conected with improving computer systems used in evaluation of skin lesions.

17

Diagnosing skin melanoma: current versus future directions

Hippe Z. S., Bajcar S., Blajdo P., Grzymała-Busse J. P., Grzymała-Busse J. W., Knap M., Paja W., Wrzesień M.

TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk

|

2003

|

Vol. 7, No 2

289-293

EN

A new database containing 410 cases of nevi pigmentosi, in four categories: benign nevus, blue nevus, suspicious nevus and melanoma malignant, carefully verified by histopathology, is described. The database is entirely different from the base presented previously, and can be readily used for research based on the so-called constructive induction in machine learning. To achieve this, the database features a different set of thirteen descriptive attributes, with a fourteenth additional attribute computed by applying values of the remaining thirteen attributes. In addition, a new program environment for the validation of computer-assisted diagnosis of melanoma, is briefly discussed. Finally, results are presented on determining optimal coefficients for the well-known ABCD formula, useful for melanoma diagnosis.

18

Application of covering algorithm for classification of melanoma spots on the skin

Grzymała-Busse J. W., Hippe Z. S.

Journal of Applied Computer Science

|

2002

|

Vol. 10, nr 2

27-33

EN

An in-house elaborated and implemented covering algorithm was applied. For the development of learning models (in the form of production rules), used then for computer-assisted classification (hence, diagnosing) of melanoma spots on the skin. In our research, four types of marks (namely, Benign nevus, Blue nevus, Suspicious nevus, and Melanoma malignant) have be en investigated. One of the generated learning models (the most promimissing one) was optimized by execution of selected generic operations on production rules, what lead to a very concise set of rules (4 rules only) giving errorless classification of unseen cases tested.