Brain Tumor Segmentation Based on Deep Learning’s Feature Representation

Brain tumor is considered as one of the most serious causes of death in the world. Thus, it is very important to detect it as early as possible. In order to predict and segment the tumor, many approaches have been proposed. However, they suffer from different problems such as the necessity of the intervention of a specialist, the long required run-time and the choice of the appropriate feature extractor. To address these issues, we proposed an approach based on convolution neural network architecture aiming at predicting and segmenting simultaneously a cerebral tumor. The proposal was divided into two phases. Firstly, aiming at avoiding the use of the labeled image that implies a subject intervention of the specialist, we used a simple binary annotation that reflects the existence of the tumor or not. Secondly, the prepared image data were fed into our deep learning model in which the final classification was obtained; if the classification indicated the existence of the tumor, the brain tumor was segmented based on the feature representations generated by the convolutional neural network architectures. The proposed method was trained on the BraTS 2017 dataset with different types of gliomas. The achieved results show the performance of the proposed approach in terms of accuracy, precision, recall and Dice similarity coefficient. Our model showed an accuracy of 91% in tumor classification and a Dice similarity coefficient of 82.35% in tumor segmentation.

Co-Encapsulation of Fisetin and Cisplatin into Liposomes for Glioma Therapy: From Formulation to Cell Evaluation

(1) Background: Glioblastoma (GBM) is the most frequent cerebral tumor. It almost always relapses and there is no validated treatment for second-line GBM. We proposed the coencapsulation of fisetin and cisplatin into liposomes, aiming to (i) obtain a synergistic effect by combining the anti-angiogenic effect of fisetin with the cytotoxic effect of cisplatin, and (ii) administrate fisetin, highly insoluble in water. The design of a liposomal formulation able to encapsulate, retain and deliver both drugs appeared a challenge. (2) Methods: Liposomes with increasing ratios of cholesterol/DOPC were prepared and characterized in term of size, PDI and stability. The incorporation of fisetin was explored using DSC. The antiangiogneic and cytotoxic activities of the selected formulation were assayed in vitro. (3) Results: We successfully developed an optimized liposomal formulation incorporating both drugs, composed by DOPC/cholesterol/DODA-GLY-PEG2000 at a molar ratio of 75.3/20.8/3.9, with a diameter of 173 ± 8 nm (PDI = 0.12 ± 0.01) and a fisetin and cisplatin drug loading of 1.7 ± 0.3% and 0.8 ± 0.1%, respectively, with a relative stability over time. The maximum incorporation of fisetin into the bilayer was determined at 3.2% w/w. Then, the antiangiogenic activity of fisetin was maintained after encapsulation. The formulation showed an additive effect of cisplatin and fisetin on GBM cells; (4) Conclusions: The developed co-loaded formulation was able to retain the activity of fisetin, was effective against GBM cells and is promising for further in vivo experimentations.

Tumor Detection and Classification of MRI Brain Image using Support Vector Machine (SVM)

Brain is one of the most complex organs in the human body that works with billions of cells. A cerebral tumor occurs when there is an uncontrolled division of cells that form an abnormal group of cells around or within the brain. This cell group can affect the normal functioning of brain activity and can destroy healthy cells. Brain tumors are classified as benign or low-grade and malignant tumors or high-grade. Benign tumors are non-cancerous tumor and they do not spread to other tissues or organs. Malignant tumors are cancerous tissue and they can easily spread to other tissues or organs. Proposed system is to differentiate between normal brain and tumor brain (benign or malign). Also, the proposed system predicts brain tumor from MRI image classification system is based on extracting useful MRI features for diagnosing the medical MRI images. The benefits of using SVM is nevertheless of the image brightness or rotation of the MRI image, it also provides huge number of strong features that can be automatically prepared well to be suitable for MRI classification. Support Vector Machine (SVM) algorithm is used to predict the diseases accurately from MRI (Magnetic Resonance Imaging) scan images. SVM algorithm is the used for the purpose of classifying the image datasets and to predict the disease by itself for those matching the images to enhance a comprehensive set of quantitative measurements among several influential on various brain image databases.

Epidemiological profile of hospitalizations in Brazil for malignant brain neoplasm among 2017 and 2020

Introduction: The malignant brain neoplasm is the cerebral tumor constituted by primitives cells with neuronal and glial diferentiation. Objective: To evaluate the profile of hospitalizations due to malignant brain neoplasm registered in Brazil among 2017 and 2020. Methods: We performed a cross secctional study from public health system database of hospitalar morbity related to hospitalizations by year and federative unit among 2017 and 2020. Results: In the period from 2017 to 2020, Brazil registered a total of 58.177 admissions for Malignant Neoplasm of the Brain. The Southeast region had highest rate of hospitalizations in that time interval, with a total of 25,266 cases (highest rate was 2019, with 6.636 cases). South region was the second place, with 13.738 hospitalizations (highest rate was 2020, with 3.613 cases). Third, Northeast region with 11,197 hospitalizations (the highest rate was 2018, with 2,837 hospitalizations). Fourth, the Central-West region, where 5.586 were hospitalizaed (with highest number of hospitalizations in 2018, 1.469 cases). The North region has the lowest incidence, with 2,390 hospitalizations (highest rate was 2019, with 665 hospitalizations). In relation to the total result of hospitalizations, the year 2019 has the highest rate of hospitalizations (n = 14.966) and the year 2020 the lowest rate (n = 14.235). Conclusion: The Southeast region had the highest rates of hospitalizations due to malignant brain câncer and the year of 2019 had the higher prevalence of hospitalized.

RADT-34. PREDICTIVE FACTORS FOR OVERALL SURVIVAL IN SURGICAL CASES OF GLIOMATOSIS CEREBRI FROM THE NATIONAL CANCER DATABASE

Gliomatosis Cerebri (GC) is a rare, aggressive, diffusely infiltrating cerebral tumor. Prognostic indicators and management strategies are currently poorly characterized. The National Cancer Database was queried for patients with histologically confirmed GC between 2004 and 2016. Demographic, tumor, and treatment characteristics were collected, including the Charlson/Deyo score, a comorbidity index adapted from the Charleston Comorbidity Index. Allowable values for the Charlson/Deyo score are 0 (no recorded comorbidities), 1, 2, and 3+ (most severe). Factors associated with overall survival were identified via bivariate log-rank tests and multivariate stepwise Cox proportional hazards models. The query returned 108 GC patients. The median age was 60.0 years, males were predominantly affected (63%), and most patients were white (86%). While 12% of cases achieved near/gross total resection and 27% of cases achieved partial resection, most surgeries were for biopsy (61%). Treatments included radiation therapy in 64% and chemotherapy in 63% of patients. The median overall survival was 15.1 (95% confidence interval [CI]=11.1-24.8) months. On bivariate analysis, chemotherapy improved overall survival (p=0.01) while radiation therapy (p=0.07) and extent of resection (p=0.48) did not. On multivariate analysis, older patients (hazard ratio [HR]=1.07, CI=1.03-1.11, p< 0.01) and Charlson/Deyo scores of ≥1 versus 0 (HR=3.47, CI=1.40-8.60, p< 0.01) had significantly increased mortality risk following surgery. In particular, the Charlson/Deyo score is a novel prognostic factor for GC that may guide clinical and surgical decision-making for this rare, rapidly fatal tumor. Further prospective studies are warranted to clarify the effects of chemotherapy versus radiation as treatment modalities for GC.

Duration and Mortality Characteristics of Patients with Cerebral Tumors Treated in the Intensive Care Unit of Sanglah General Hospital from January to December 2017

BACKGROUND: The incidence and survival rates of cerebral tumor patients are increasing, along with advances in medical technology. Severe complications, such as multiorgan failure in patients, cause the need for intensive care to improve. The effectiveness of intensive care in tumor patients is still controversial, considering that patients included in low and vulnerable performance status. Patients with cerebral tumors also require intensive care either before or after effective action. The controversy of economic and ethical issues still places aggressive action in intensive care on oncology patients. It is necessary to identify and know the characteristics of tumor patients that are useful in intensive management. AIM: The purpose of this study was to determine the demographic characteristics, length of stay, complications, and mortality rates of cerebral tumor patients undergoing intensive care at Sanglah Hospital Denpasar, the period January–December 2017. METHODS: This study is a cross-sectional descriptive where all sample data collection is secondarily based on existing medical records. The sample population in this study were cerebral tumor patients treated in the intensive care ward Sanglah period January–December 2017 who met the inclusion criteria. RESULTS: In the study, it found that 54 patients with the highest number of cerebral tumors treated in the intense sex with women (64.85%) and middle-aged adults (41–60 years) were 31 patients (57.4%). Most diagnoses of cerebral tumors were 39 non-glioma tumors (72.2%) followed by glioma tumors (20.4%) and tumor metastases (7.4%). Non-glioma tumors noted were meningiomas of around 48.1%, followed by glioblastoma multiforme tumors of 9.3%, tumors at the cerebellopontine angle and tumor metastases of 7.4%, respectively. All patients were treated intensively after receiving surgical therapy with a Karnofsky 50 score of about 29 patients (53.7%). All patients came out alive, and there were no recorded complications during treatment with the maximum length of stay of patients in the intensive care unit for up to 3 days by 45 patients (83.3%). CONCLUSION: The most brain tumor cases are meningioma, which occurred in women, middle-aged adults with no mortality and complications within 3 days during treatment intensive care.

Intra-operative radiation therapy as salvage treatment option for recurrent glioblastoma multiforme.

2535 Background: Glioblastoma multiforme (GBM) is an extremely aggressive cerebral tumor with poor prognosis. The majority of patients relapse after the initial surgery plus adjuvant radiation and chemotherapy. In case of recurrence there is no established standard therapy. The optimal techniques for salvage re-irradiation are unclear, so that procedure poses a challenge. In contrast to traditional external beam radiotherapy (EBRT) intra-operative radiotherapy (IORT) may improve patient’s outcome at the cost of minimal side effects and short treatment duration. Methods: A total of 30 patients were treated with recurrent GBM between August 2016 and June 2019. All patients underwent maximal safe resection; patients were divided into IORT and EBRT groups. 15 patients were included in each group with similar clinical characteristics. All patients in IORT group underwent maximal safe microsurgical resection with subsequent intraoperative balloon electronic brachytherapy (IBEB) and no further adjuvant treatment. IBEB was performed using Axxent electronic brachytherapy device (Xoft Electronic Brachytherapy (eBx) System, USA. Patients in EBRT group underwent same surgery followed by external beam radiotherapy. Contrast-enhanced brain MRI with perfusion was performed within 24 hours of surgery +/- brain PET-CT with 18-FDOPA and then every 3 months. Both groups were also assigned to subgroups (≤ 2.5cm3 and > 2.5cm3) based on post-operative contrast-enhancing volume (POCEV). Median overall survival (OS) since diagnosis and local progression-free survival (locPFS) following the second surgery were analyzed. Possible toxicities and prognostic factors were also evaluated. Results: Median OS was 27 months in IORT group and 21 months in EBRT group. The locPFS range between 3.5 to 39 months in IORT group and only 2 to 10 months in group with EBRT. Kaplan-Meier OS curves in patients with POCEV ≤ 2.5cm3 showed more favorable outcomes for patients in the IORT group (p < 0.05). In patients with POCEV > 2.5cm3 the median OS was 17 months in IORT group and 13.5 months in EBRT group. Conclusions: IORT of recurrent GBM is feasible and provides encouraging local progression-free and overall survival; no high-grade radiation induced toxicities occur and further studies to establish this method are mandatory. The toxicity profile of additional IBEB was manageable. Maximal safe microsurgical resection is the most important prognostic factor and could determine the effectiveness of post-surgical IBEB.