Damian Borys

Integrated System Supporting Research on Environment Related Cancers

There are many impediments to progress in cancer research. Insufficient or low quality data and computational tools that are dispersed among various sites are one of them. In this paper we present an integrated system that combines all stages of cancer studies, from gathering of clinical data, through elaborate patient questionnaires and bioinformatics tools, to data warehousing and preparation of analysis reports.

A Holistic Approach to Testing Biomedical Hypotheses and Analysis of Biomedical Data

Testing biomedical hypotheses is performed based on advanced and usually many-step analysis of biomedical data. This requires sophisticated analytical methods and data structures that allow to store intermediate results, which are needed in the subsequent steps. However, biomedical data, especially reference data, often change in time and new analytical methods are created every year. This causes the necessity to repeat the iterative analyses with new methods and new reference data sets, which in turn causes frequent changes of the underlying data structures. Such instability of data structures can be mitigated by the use of the idea of data lake, instead of traditional database systems.

Scalability of a Genomic Data Analysis in the BioTest Platform

BioTest platform is dedicated for the processing of biomedical data that originate from various measurement techniques. This includes next-generation sequencing (NGS), that focuses the attention of researchers all of the world due to its broad possibilities in determining the structure of the DNA and RNA. However, the analysis of data provided by NGS requires large disk space, and is time-consuming, becoming a challenge for the data processing systems. In this paper, we have analyzed the possibility of scaling the BioTest platform in terms of genomic data analysis and platform architecture. Scalability tests were carried out using next-generation sequencing data and relied on methods for detection of somatic mutations and polymorphisms in the human DNA. Our results show that the platform is scalable, allowing to significantly reduce the execution time of performed calculations. However, the scalability capabilities depend on the experiment methodology and homogeneity of resources required by each task, which in NGS studies can be highly variable.

Automatic Segmentation System of Emission Tomography Data Based on Classification System

Segmentation and delineation of tumour boundaries are important and difficult step in emission tomography imaging, where acquired and reconstructed images presents large noise and a blurring level. Several methods have been previously proposed and can be used in single photon emission tomography (SPECT) or positron emission tomography (PET) imaging. Some of them relies on the standard uptake value (SUV) used in PET imaging. Presented approach can be used in both (SPECT and PET) modalities and it is based on support vector machines (SVM) classification system. System has been tested on standard phantom, widely used for testing the emission tomography devices. Results are presented for two classifiers SVM and DLDA.

Breast Deformation Modeling Based on MRI Images, Preliminary Results

Increasing number of people are suffering from cancer these days. The highest incidence among women is breast cancer. One of the main sources of diagnostic information is Magnetic Resonance Imaging (MRI). Another one is Positron Emission Tomography with Computer Thomography (PET-CT). Both examinations take place in different patient positions (prone and supine respectively). The only way to obtain complete diagnostic information is to bring MRI images into PET-CT space and compare both. Our preliminary studies focus on creating a simple, deformable finite element model of the breast. Proposed algorithm allows to obtain breast model in supine position, created from MRI images obtained in prone position.

A Simple Hair Removal Algorithm from Dermoscopic Images

The main goal of our work is initial preprocessing of dermoscopic images by hair removal. Dermoscopy is a basic technique in skin melanoma diagnostics. One of the main problems in dermoscopy images analysis are hairs objects in the image. Hairs partially shade the main region of interest that’s why it needs special treatment. We have developed a simple and fast hair removal algorithm based on basic image processing algorithms. The algorithm was tested on available online test database PH2 [6]. Primary results of proposed algorithm show that even if hair contamination in the image is significant algorithm can find those objects. There is still a place for improvements as long as some air bubbles are marked as a region of interest.

The evaluation of correction algorithms of intensity nonuniformity in breast MRI images: a phantom study

The aim of this work was to test the most popular and essential algorithms of the intensity nonuniformity correction of the breast MRI imaging. In this type of MRI imaging, especially in the proximity of the coil, the signal is strong but also can produce some inhomogeneities. Evaluated methods of signal correction were: N3, N3FCM, N4, Nonparametric, and SPM. For testing purposes, a uniform phantom object was used to obtain test images using breast imaging MRI coil. To quantify the results, two measures were used: integral uniformity and standard deviation. For each algorithm minimum, average and maximum values of both evaluation factors have been calculated using the binary mask created for the phantom. In the result, two methods obtained the lowest values in these measures: N3FCM and N4, however, for the second method visually phantom was the most uniform after correction.

Algorithm for Simple Automated Breast MRI Deformation Modelling.

Increasing incidence of breast cancer caused development of patient-specific treatment planning procedures. The most effective tool for breast cancer visualisation is Magnetic Resonance Imaging (MRI). However, the MRI scans represent patient data in prone position with breast placed in signal enhancement coils, while other procedures, i.e. surgery, PET-CT (Positron Emission Tomography fused with Computer Tomography) are performed in patient supine position. The bigger patient breast is, the bigger its shape differ in every patient position, what influences its interior structure and tumour location. In this paper, we present our method for automated breast model deformation, which is based on prone MRI dataset. Proposed algorithm allows to obtain reliable breast model in supine position in a few simple steps, without manual intervention.

Rigid and Non-rigid Registration Algorithm Evaluation in MRI for Breast Cancer Therapy Monitoring

One of the most common methods in breast cancer radiotherapy planning is Magnetic Resonance Imaging (MRI). It is also used for patient evaluation during treatment because of its sensitivity and lack of ionizing radiation. During each imaging session a patient position can be different and inaccuracies can occur. In this case it is very difficult to compare two image sets originating from different patient examination. The main goals of this work were to implement an algorithm, based on affine transformation with Mutual Information as the quality factor of images match and the method based on the Navier-Lame equation for elastic image co-registration. The rigid transformation is used for the preliminary processing, and the non-rigid transformation allows for successful co-registration of both image sets. Our results were evaluated visually, and the MI indices were calculated. These algorithms allowed for image co-registration in different imaging sessions during the course of treatment.

Games with Resources and Their Use in Modeling Effects of Anticancer Treatment

In this work, we study an extension to evolutionary game models with the possibility to model the change and influence of the environment fluctuations for the fitness of the players. Using spatial games, those changes can be taken into account as an additional lattice dimension. In classical spatial evolutionary games (SEGT) each position on the lattice is represented by a single player or single phenotype (strategy). The local payoff for this player arises from the interactions with the neighbouring cells. With the newer approach each cell represents heterogeneous subpopulation, so can be considered as mixed or multidimensional spatial games (MSEG). This allows performing the game on a multidimensional lattice where an additional dimension is representing the evolution of resources.