|
|
The diagnostic and prognostic ability of selected serum and urinary markers of prostate cancer
Do Carmo Silva, Joana Isabel ; Veselý, Štěpán (advisor) ; Čapoun, Otakar (referee) ; Fedorko, Michal (referee)
The diagnostic and prognostic ability of selected serum and urinary markers of prostate cancer Abstract Serum prostate specific antigen (PSA) is the only widely approved marker in prostate cancer (PC) diagnosis and follow up after treatment. Its role has remained controversial due to lack of specificity and the risk of overdiagnosis of insignificant PC. The aim of this work was to explore promising markers of PC and to improve current patient stratification to adjuvant treatment. Three main studies were performed using different media (urine and serum). The first study included the evaluation of Engrailed-2 (EN2) - a urinary marker of interest - in 90 patients with localized PC, 30 healthy controls, and 40 patients indicated for prostate biopsy. The second study evaluated 205 men with high-risk PC-features who underwent radical prostatectomy (RP) and were subject to a strict follow-up protocol of ultrasensitive PSA (UPSA) at close time intervals. The ability of particular measurements to predict biochemical recurrence (BCR) and thus the need for adjuvant therapy was assessed using the area under the curve (AUC) and a stratification model was created. The third study involved 128 patients who underwent RP. PSA and its serum isoforms normally used in the diagnostic context were evaluated both preoperatively...
|
|
|
Chromophobe Renal Cell Cancer
Kolář, Jiří ; Hora, Milan (advisor) ; Pivovarčíková, Kristýna (referee) ; Fedorko, Michal (referee)
Introduction: Chromophobe renal cell carcinoma (chRCC) is the third most common subtype of RCC after clear cell RCC and papillary RCC. It is a relatively indolent subtype of RCC with approximately 5% risk of progression and development of metastases after surgical treatment. Predicting the clinical behaviour of chRCC by histologic features has so far proven to be challenging which is related to the lack of a validated grading system for this type of renal cancer. Historically, several grading schemes were proposed specifically for chRCC. However, none of them has proven useful in clinical practice. Two main histological subtypes of chRCC are known, namely classic and eosinophilic. Moreover, several aberrant histologic subtypes have been described, including adenomatoid microcystic pigmented, multicystic, neuroendocrine, papillary, oncocytic and small cell-like. It has been previously reported that both main subtypes of chRCC (classic and eosinophilic) have no differences in their prognosis. It is hypothesized that also these rare histological subtypes are not associated with differences in clinical outcomes. Nevertheless, this hypothesis has never been tested or confirmed by clinical data. Materials and methods: The aim of this thesis was to assess the impact of histologic diversity in chRCC...
|
|
|
Rare Tumours of the Kidney - Cystic Tumours of the Kidney
Pitra, Tomáš ; Hora, Milan (advisor) ; Fedorko, Michal (referee) ; Weibl, Peter (referee)
Aim: To quantify the occurrence of cystic tumours of the kidney in a population of patients with tumour of the kidney. Other goals of this study were to compare results of imaging studies with the results of final histology, to evaluate malignity rate in each Bosniak group and to evaluate the options of surgical treatment of cystic tumours of the kidney. The emphasis was placed to determine the benefits of magnetic resonance imaging in the diagnostic algorithm of cystic tumours of the kidney. Material and methods: All patients, who underwent surgery for the tumour of the kidney or cystic lesion of the kidney at our department in the time period 2009-17 were evaluated and patients with radiologically described cystic lesion of the kidney were included in the cohort. Lesions were classified according to CT imaging in the Bosniak categories. The unclear findings were in clinical study consulted with the radiologist. In case of uncertainties, MRI was indicated. Histological classification of all neoplasias was done according to the WHO classification 2016. The cohort of cystic lesions (n=247) was stratified into two groups. Categories Bosniak I-IIF (n=115), which were treated surgically not for suspicion of the malignity, but because of the size, localisation, clinical manifestation or were solved...
|
|
|
Rare Tumours of the Kidney - Cystic Tumours of the Kidney
Pitra, Tomáš ; Hora, Milan (advisor) ; Fedorko, Michal (referee) ; Weibl, Peter (referee)
Aim: To quantify the occurrence of cystic tumours of the kidney in a population of patients with tumour of the kidney. Other goals of this study were to compare results of imaging studies with the results of final histology, to evaluate malignity rate in each Bosniak group and to evaluate the options of surgical treatment of cystic tumours of the kidney. The emphasis was placed to determine the benefits of magnetic resonance imaging in the diagnostic algorithm of cystic tumours of the kidney. Material and methods: All patients, who underwent surgery for the tumour of the kidney or cystic lesion of the kidney at our department in the time period 2009-17 were evaluated and patients with radiologically described cystic lesion of the kidney were included in the cohort. Lesions were classified according to CT imaging in the Bosniak categories. The unclear findings were in clinical study consulted with the radiologist. In case of uncertainties, MRI was indicated. Histological classification of all neoplasias was done according to the WHO classification 2016. The cohort of cystic lesions (n=247) was stratified into two groups. Categories Bosniak I-IIF (n=115), which were treated surgically not for suspicion of the malignity, but because of the size, localisation, clinical manifestation or were solved...
|
|
|
Datamining in data from financial institution
Fedorko, Michal ; Rauch, Jan (advisor) ; Kotlář, Ondřej (referee)
The main purpose of this paper is to create datamining analysys of volutary termations in financial institution based in Czech republic on the data stored by HR department. Only hard data currently stored was input. For creating analysys the CRISP-DM metodology was used. For modeling itself the LISp-Miner was used. Association rules were the main approach to solving the task..Several interesting association rules were found and interprated. Outcome of the paper is for internal campains of the customer and there is even motivation for furthure predictive modeling and this was the first step.
|
guest ::
RSS feed.