ABSTRACT
Background. It is estimated that 347 million people suffer from diabetes mellitus (DM), and almost 5 million are blind due to diabetic retinopathy (DR). The progression of DR can be slowed down with early diagnosis and treatment. Therefore our aim was to develop a novel automated method for DR screening. Methods. 52 patients with diabetes mellitus were enrolled into the project. Of all patients, 39 had signs of DR. Digital retina images and tear fluid samples were taken from each eye. The results from the tear fluid proteomics analysis and from digital microaneurysm (MA) detection on fundus images were used as the input of a machine learning system. Results. MA detection method alone resulted in 0.84 sensitivity and 0.81 specificity. Using the proteomics data for analysis 0.87 sensitivity and 0.68 specificity values were achieved. The combined data analysis integrated the features of the proteomics data along with the number of detected MAs in the associated image and achieved sensitivity/specificity values of 0.93/0.78. Conclusions. As the two different types of data represent independent and complementary information on the outcome, the combined model resulted in a reliable screening method that is comparable to the requirements of DR screening programs applied in clinical routine.
Subject(s)
Aneurysm/diagnosis , Diabetes Mellitus/metabolism , Diabetic Retinopathy/diagnosis , Fundus Oculi , Proteome/metabolism , Retina , Retinal Vessels , Tears/metabolism , Aged , Biomarkers/metabolism , Case-Control Studies , Female , Humans , Image Processing, Computer-Assisted , Machine Learning , Male , Mass Screening , Middle Aged , Photography , Proteomics , Sensitivity and SpecificityABSTRACT
BACKGROUND: The aim of the project was to develop a novel method for diabetic retinopathy screening based on the examination of tear fluid biomarker changes. In order to evaluate the usability of protein biomarkers for pre-screening purposes several different approaches were used, including machine learning algorithms. METHODS: All persons involved in the study had diabetes. Diabetic retinopathy (DR) was diagnosed by capturing 7-field fundus images, evaluated by two independent ophthalmologists. 165 eyes were examined (from 119 patients), 55 were diagnosed healthy and 110 images showed signs of DR. Tear samples were taken from all eyes and state-of-the-art nano-HPLC coupled ESI-MS/MS mass spectrometry protein identification was performed on all samples. Applicability of protein biomarkers was evaluated by six different optimally parameterized machine learning algorithms: Support Vector Machine, Recursive Partitioning, Random Forest, Naive Bayes, Logistic Regression, K-Nearest Neighbor. RESULTS: Out of the six investigated machine learning algorithms the result of Recursive Partitioning proved to be the most accurate. The performance of the system realizing the above algorithm reached 74% sensitivity and 48% specificity. CONCLUSIONS: Protein biomarkers selected and classified with machine learning algorithms alone are at present not recommended for screening purposes because of low specificity and sensitivity values. This tool can be potentially used to improve the results of image processing methods as a complementary tool in automatic or semiautomatic systems.
Subject(s)
Diabetic Retinopathy/diagnosis , Eye Proteins/metabolism , Tears/metabolism , Adult , Algorithms , Biomarkers/metabolism , Diabetic Retinopathy/metabolism , Female , Humans , Logistic Models , Male , Sensitivity and Specificity , Tandem Mass SpectrometryABSTRACT
UNLABELLED: Biobanks are essential tools in diagnostics and therapeutics research and development related to personalized medicine. Several international recommendations, standards and guidelines exist that discuss the legal, ethical, technological, and management requirements of biobanks. Today's biobanks are much more than just collections of biospecimens. They also store a huge amount of data related to biological samples which can be either clinical data or data coming from biochemical experiments. A well-designed biobank software system also provides the possibility of finding associations between stored elements. Modern research biobanks are able to manage multicenter sample collections while fulfilling all requirements of data protection and security. While developing several biobanks and analyzing the data stored in them, our research group recognized the need for a well-organized, easy-to-check requirements guideline that can be used to develop biobank software systems. International best practices along with relevant ICT standards were integrated into a comprehensive guideline: The Model Requirements for the Management of Biological Repositories (BioReq), which covers the full range of activities related to biobank development. The guideline is freely available on the Internet for the research community. AVAILABILITY: The database is available for free at http://bioreq.astridbio.com/bioreq_v2.0.pdf.
ABSTRACT
UNLABELLED: The ever evolving Next Generation Sequencing technology is calling for new and innovative ways of data processing and visualization. Following a detailed survey of the current needs of researchers and service providers, the authors have developed GenoViewer: a highly user-friendly, easy-to-operate SAM/BAM viewer and aligner tool. GenoViewer enables fast and efficient NGS assembly browsing, analysis and read mapping. It is highly customized, making it suitable for a wide range of NGS related tasks. Due to its relatively simple architecture, it is easy to add specialised visualization functionalities, facilitating further customised data analysis. The software's source code is freely available; it is open for project and task-specific modifications. AVAILABILITY: The database is available for free at http://www.genoviewer.com/
