Spatiotemporal Distribution, Time to Treatment Outcome Clustering and Determinants of Esophageal Cancer in Ethiopia, a Scoping Study.

INTRODUCTION: Esophageal cancer was the eighth and sixth leading cause of morbidity of all cancers in the world, and the 15th and 12th in Ethiopia, respectively. There is a lack of comprehensive data regarding Ethiopia's esophageal cancer hotspot, treatment outcome clustering, and other factors. OBJECTIVE: This scoping review was designed to understand the extent and type of existing evidence regarding spatiotemporal distribution, time to treatment outcome clustering, and determinants of esophageal cancer in Ethiopia up to March 28, 2023. METHODS: Three-step search strategies were employed for the scoping review from March 15 to 28, 2023. Targeted databases included PubMed/Medline, PubMed Central (PMC), Google Scholar, Hinari, and Cochrane for published studies and different websites for unpublished studies for evidence synthesis. Data were extracted using the Joanna Briggs Institute (JBI) manual format. RESULTS: Our final analysis comprised 17 (16 quantitative and 1 qualitative) studies. Three studies attempted to depict the country's temporal distribution, whereas 12 studies showed the spatial distribution of esophageal cancer by proportion. The regional state of Oromia recorded a high percentage of cases. Numerous risk factors linked to the tumor have been identified in 8 investigations. Similarly, 5 studies went into detail regarding the likelihood of survival and the factors that contribute to malignancy, while 2 studies covered the results of disease-related treatments. CONCLUSIONS: The substantial body of data that underpins this finding supports the fact that esophageal cancer has several risk factors and that its prevalence varies greatly across the country and among regions. Surgery, radiotherapy, or chemotherapy helped the patient live longer. However, no research has investigated which treatment is best for boosting patient survival and survival clustering. Therefore, research with robust models for regional distribution, clustering of time to treatment outcomes, and drivers of esophageal cancer will be needed.

The review was based on 17 studies searched from five electronic databases, and six additional sources. Esophageal cancer incidence varies across the nation (from region to region). The median survival time of esophageal cancer cases were four months, and six months. No study investigated the better treatment that improved the survival of patients with esophageal cancer. A contradicting report were found about the link b/n khat chewing and esophageal cancer. The temporal distribution of the tumor was controversial.

Determinants of COVID-19 severity in Ethiopia: a multicentre case-control study.

OBJECTIVE: It was necessary to understand the determinants of severe COVID-19 in order to deliver targeted healthcare services to prevent further complications and mortality. Identifying the factors associated with severe COVID-19 in Addis Ababa, Ethiopia, is the aim of this study. DESIGN: A case-control study was conducted from October 2021 to March 2022. SETTING: The study was conducted at three public COVID-19 treatment centres including Ekka Kotebe General, St. Peter Hospital and St. Paul's Hospital. PARTICIPANTS: The study participants were COVID-19 patients admitted to three COVID-19 treatment centres. Cases were patients admitted with severe COVID-19, and controls were patients with mild or moderate COVID-19. A total of 306 patients (153 cases and 153 controls) selected by simple random sampling technique participated in this study. OUTCOME MEASURES AND ANALYSIS: Data were collected by a face-to-face or telephone interviewer using a structured questionnaire. COVID-19 admission category, clinical and biomedical characteristics and comorbidity-related data were extracted from the participant's medical record. Multivariable binary logistic regression analysis was used to identify predictors of COVID-19 severity. RESULTS: The odds of being old were 4.54 times higher among severe COVID-19 cases (adjusted odds ratio (AOR)=4.54, 95% CI=2.499 to 8.24), the odds of being male were 2.72 times higher among severe COVID-19 cases (AOR=2.72, 95% CI=1.46 to 5.057), being vaccinated for COVID-19 decreases the severity by 55.1% (AOR=0.449, 95% CI=0.251 to 0.801), having good knowledge about COVID-19 decreases by 65% (AOR=0.35%, 95% CI=0.195 to 0.63) among patients with severe COVID-19, the odds of being diabetic were 10.2 times higher among severe COVID-19 cases (AOR=10.2, 95% CI=4.596 to 22.61) and the odds of being hypertensive were 2.3 times higher among severe COVID-19 cases (AOR=2.26, 95% CI=1.092 to 4.685). CONCLUSION: Male, older age, diabetes or hypertension comorbidity, COVID-19 vaccination and having inadequate knowledge about COVID-19 were determinant factors of severe COVID-19.

Magnitude of standard precautions practices among healthcare workers in health facilities of Low and Middle Income Countries: A systematic review and meta-analysis.

BACKGROUND: Standard precautions are the minimum standard of infection control to prevent transmission of infectious agents, protect healthcare workers, patients, and visitors regardless of infection status. The consistent implementation of standard precautions is highly effective in reducing transmission of pathogens that cause HAIs. Despite their effectiveness, compliance, resources, patient behavior, and time constraints are some of the challenges that can arise when implementing standard precautions. The main objective of this meta-analysis was to show the pooled prevalence of safe standard precaution practices among healthcare workers in Low and Middle Income Countries (LMICs). METHODS: A systematic review and meta-analysis was conducted for this study. We systematically searched observational study articles from PubMed Central and Google Scholar. We included articles published any year and involving healthcare workers. We used Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). The random effect model was used to estimate the pooled prevalence. The meta-analysis, sensitivity analysis, subgroup analysis, and publication bias (funnel plot, and Egger's tests) were conducted. RESULTS: A total of 46 articles were included in this study. The pooled prevalence of standard precautions practices among healthcare workers in LMICs was 53%, with a 95% CI of (47, 59). These studies had a total sample size of 14061 with a minimum sample size of 17 and a maximum sample size of 2086. The majority of the studies (82.6%) were conducted in hospitals only (all kinds), and the remaining 17.4% were conducted in all health facilities, including hospitals. CONCLUSIONS: The pooled prevalence of standard precautions practices among healthcare workers in LMICs was suboptimal. The findings of this study can have substantial implication for healthcare practice and policy making by providing robust evidence with synthesized and pooled evidence from multiple studies. TRIAL REGISTRATION: Registered on PROSPERO with record ID: CRD42023395129, on the 9th Feb. 2023.

Dual view deep learning for enhanced breast cancer screening using mammography.

Breast cancer has the highest incidence rate among women in Ethiopia compared to other types of cancer. Unfortunately, many cases are detected at a stage where a cure is delayed or not possible. To address this issue, mammography-based screening is widely accepted as an effective technique for early detection. However, the interpretation of mammography images requires experienced radiologists in breast imaging, a resource that is limited in Ethiopia. In this research, we have developed a model to assist radiologists in mass screening for breast abnormalities and prioritizing patients. Our approach combines an ensemble of EfficientNet-based classifiers with YOLOv5, a suspicious mass detection method, to identify abnormalities. The inclusion of YOLOv5 detection is crucial in providing explanations for classifier predictions and improving sensitivity, particularly when the classifier fails to detect abnormalities. To further enhance the screening process, we have also incorporated an abnormality detection model. The classifier model achieves an F1-score of 0.87 and a sensitivity of 0.82. With the addition of suspicious mass detection, sensitivity increases to 0.89, albeit at the expense of a slightly lower F1-score of 0.79.

Infection prevention and control practices in the Pediatrics and Child Health Department of Tikur Anbessa Specialized Hospital, Ethiopia.

Background: Infection prevention and control (IPC) is a set of practices that are designed to minimize the risk of healthcare-associated infections (HAIs) spreading among patients, healthcare workers, and visitors. Implementation of IPC is essential for reducing infection incidences, preventing antibiotic use, and minimizing antimicrobial resistance (AMR). The aim of the study was to assess IPC practices and associated factors in Pediatrics and Child Health at Tikur Anbessa Specialized Hospital. Methods: In this study, we used a cross-sectional study design with a simple random sampling method. We determined the sample size using a single population proportion formula with the assumption of a 55% good IPC practice, a 5% accepted margin of error, and a 15% non-response rate and adjusted with the correction formula. The final sample size was 284 healthcare workers. The binary logistic regression model was used for analysis. The World Health Organization (WHO) Infection Prevention and Control Assessment Framework (IPCAF) tool was used to assess IPC core components. Result: A total of 272 healthcare workers participated in the study, with a response rate of 96%. Of the total participants, 65.8% were female and 75.7% were nurses. The overall composite score showed that the prevalence of good IPC practices among healthcare workers was 50.4% (95% CI: 44.3-56.5). The final model revealed that nursing professionals and healthcare workers who received IPC training had AORs of 2.84 (95% CI: 1.34-6.05) and 2.48 (95% CI: 1.36-4.52), respectively. The final average total IPCAF score for the IPC level was 247.5 out of 800 points. Conclusion: The prevalence of good IPC practice was suboptimal. The study participants, who were nursing professionals and healthcare workers who received IPC training, showed a statistically significant association with the IPC practice level. The facility-level IPCAF result showed a "Basic" level of practice based on the WHO categorization. These evidences can inform healthcare workers and decision-makers to identify areas for improvement in IPC practice at all levels. Training of healthcare workers and effective implementation of the eight IPC core components should be strengthened to improve suboptimal practices.

Gap analysis of service quality and associated factors at the oncology center of Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia, 2022: a cross-sectional study.

OBJECTIVE: To measure the gap between expectations of patients with cancer for oncology services and their perceptions of the actual service and to identify associated factors at the oncology centre of Tikur Anbessa Specialized Hospital, Ethiopia. DESIGN: An institutional-based cross-sectional study design was conducted using the service quality (SERVQUAL) tool from March to April 2022 on a sample of 256 hospitalised patients with cancer at the oncology centre of Tikur Anbessa Specialized Hospital. A paired Wilcoxon test and Kruskal-Wallis tests were used to determine the statistically significant difference between expectation and perception and to quantify the strength of association between the level of gap in the quality of oncology service and dependent variables, respectively. RESULTS: Out of 256 patients with cancer included in the study, all of them agreed and participated, making the response rate 100%. The overall gap in service quality explained by the mean and SD is -1.42 (±0.41). The overall score for expectation and perception is 4.24 (±0.31) and 2.82 (±0.37), respectively. Being female, age greater than 65, having a college degree and above, being a patient with cervical cancer, patients with stage 4 cancer and patients who waited for more than 12 months for radiotherapy were found to have a statistically significant higher expectation compared with their perceived care in one or more dimensions of the SERVQUAL tool. CONCLUSION: Patient perceptions of the quality of service they received were lower than their expectations of the quality of service in all service quality aspects at Tikur Anbessa Specialized Hospital's oncology centre, implying unmet quality expectations from the oncology service users, with tangibility, assurance and empathy being the dimensions with the highest gap recorded, respectively. Therefore, the hospital and other stakeholders should strive to exceed patient expectations and the overall quality of care.

Skin Lesion Classification and Detection Using Machine Learning Techniques: A Systematic Review.

Skin lesions are essential for the early detection and management of a number of dermatological disorders. Learning-based methods for skin lesion analysis have drawn much attention lately because of improvements in computer vision and machine learning techniques. A review of the most-recent methods for skin lesion classification, segmentation, and detection is presented in this survey paper. The significance of skin lesion analysis in healthcare and the difficulties of physical inspection are discussed in this survey paper. The review of state-of-the-art papers targeting skin lesion classification is then covered in depth with the goal of correctly identifying the type of skin lesion from dermoscopic, macroscopic, and other lesion image formats. The contribution and limitations of various techniques used in the selected study papers, including deep learning architectures and conventional machine learning methods, are examined. The survey then looks into study papers focused on skin lesion segmentation and detection techniques that aimed to identify the precise borders of skin lesions and classify them accordingly. These techniques make it easier to conduct subsequent analyses and allow for precise measurements and quantitative evaluations. The survey paper discusses well-known segmentation algorithms, including deep-learning-based, graph-based, and region-based ones. The difficulties, datasets, and evaluation metrics particular to skin lesion segmentation are also discussed. Throughout the survey, notable datasets, benchmark challenges, and evaluation metrics relevant to skin lesion analysis are highlighted, providing a comprehensive overview of the field. The paper concludes with a summary of the major trends, challenges, and potential future directions in skin lesion classification, segmentation, and detection, aiming to inspire further advancements in this critical domain of dermatological research.

Weak Localization of Radiographic Manifestations in Pulmonary Tuberculosis from Chest X-ray: A Systematic Review.

Pulmonary tuberculosis (PTB) is a bacterial infection that affects the lung. PTB remains one of the infectious diseases with the highest global mortalities. Chest radiography is a technique that is often employed in the diagnosis of PTB. Radiologists identify the severity and stage of PTB by inspecting radiographic features in the patient's chest X-ray (CXR). The most common radiographic features seen on CXRs include cavitation, consolidation, masses, pleural effusion, calcification, and nodules. Identifying these CXR features will help physicians in diagnosing a patient. However, identifying these radiographic features for intricate disorders is challenging, and the accuracy depends on the radiologist's experience and level of expertise. So, researchers have proposed deep learning (DL) techniques to detect and mark areas of tuberculosis infection in CXRs. DL models have been proposed in the literature because of their inherent capacity to detect diseases and segment the manifestation regions from medical images. However, fully supervised semantic segmentation requires several pixel-by-pixel labeled images. The annotation of such a large amount of data by trained physicians has some challenges. First, the annotation requires a significant amount of time. Second, the cost of hiring trained physicians is expensive. In addition, the subjectivity of medical data poses a difficulty in having standardized annotation. As a result, there is increasing interest in weak localization techniques. Therefore, in this review, we identify methods employed in the weakly supervised segmentation and localization of radiographic manifestations of pulmonary tuberculosis from chest X-rays. First, we identify the most commonly used public chest X-ray datasets for tuberculosis identification. Following that, we discuss the approaches for weakly localizing tuberculosis radiographic manifestations in chest X-rays. The weakly supervised localization of PTB can highlight the region of the chest X-ray image that contributed the most to the DL model's classification output and help pinpoint the diseased area. Finally, we discuss the limitations and challenges of weakly supervised techniques in localizing TB manifestations regions in chest X-ray images.

Prevalence and types of anemia among people with tuberculosis in Africa: a systematic review and meta-analysis.

Globally, tuberculosis (TB) and anemia are public health problems related with high morbidity and mortality. Furthermore, anemia is frequently manifested among people with TB in Africa, prevalence ranging from 25 to 99%. The presence of anemia is associated with an increase in individuals' susceptibility to TB and poor treatment outcomes. Studies have reported heterogeneous estimate of prevalence of anemia among people with TB in Africa. This review aimed to estimate the prevalence of anemia among newly diagnosed people with TB n Africa. We searched studies in Medline/PubMed, Cochrane library, ScienceDirect, JBI database, the Web of Science, Google Scholar, WorldCat, Open Grey, Scopus, Agency for Healthcare Research and Quality, ProQuest, and African Journals Online that reported the prevalence of anemia at TB diagnosis. Two reviewers performed data extraction with pre-defined inclusion criteria. A random-effects logistic regression model was used to pool the prevalence of anemia and levels of anemia with a 95% confidence interval (CI) in STATA version 14. Heterogeneity and publication biases were explored. A total of 1408 studies were initially identified, and seventeen studies with 4555 people with TB were included in the analysis. The prevalence of anemia among people with TB in Africa was 69% (95% CI 60.57-77.51). The pooled prevalence of anemia of chronic disease was 48% (95% CI 13.31-82.75) and normocytic normochromic anemia was 32% (95% CI 13.74-50.94) while mild anemia was 34% (95% CI 20.44-46.86). Females were more anemic than males at TB diagnosis in Africa (74% vs. 66%). The finding indicates that anemia is a common co-morbidity present among people with TB, especially among females. Mild anemia and normocytic normochromic anemia were more common at TB diagnosis. The finding indicates that anemia is a common co-morbidity present among people with TB in Africa region. Hence, it is recommended to instigate a routine anemia screening at TB diagnosis to improve treatment outcomes.

Review of Botnet Attack Detection in SDN-Enabled IoT Using Machine Learning.

The orchestration of software-defined networks (SDN) and the internet of things (IoT) has revolutionized the computing fields. These include the broad spectrum of connectivity to sensors and electronic appliances beyond standard computing devices. However, these networks are still vulnerable to botnet attacks such as distributed denial of service, network probing, backdoors, information stealing, and phishing attacks. These attacks can disrupt and sometimes cause irreversible damage to several sectors of the economy. As a result, several machine learning-based solutions have been proposed to improve the real-time detection of botnet attacks in SDN-enabled IoT networks. The aim of this review is to investigate research studies that applied machine learning techniques for deterring botnet attacks in SDN-enabled IoT networks. Initially the first major botnet attacks in SDN-IoT networks have been thoroughly discussed. Secondly a commonly used machine learning techniques for detecting and mitigating botnet attacks in SDN-IoT networks are discussed. Finally, the performance of these machine learning techniques in detecting and mitigating botnet attacks is presented in terms of commonly used machine learning models' performance metrics. Both classical machine learning (ML) and deep learning (DL) techniques have comparable performance in botnet attack detection. However, the classical ML techniques require extensive feature engineering to achieve optimal features for efficient botnet attack detection. Besides, they fall short of detecting unforeseen botnet attacks. Furthermore, timely detection, real-time monitoring, and adaptability to new types of attacks are still challenging tasks in classical ML techniques. These are mainly because classical machine learning techniques use signatures of the already known malware both in training and after deployment.

Association of risk factors and breast cancer among women treated at Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia: a case-control study.

OBJECTIVES: Many factors known to increase the risk of breast cancer, such as age, family history, early menarche and late menopause are not modifiable. Modifiable factors include obesity, use of menopausal hormones and breast feeding. This study aimed to assess risk factors associated with breast cancer among women at Tikur Anbessa Specialized Hospital. DESIGN: Facility based case-control study. METHODS: Case-control study was conducted from May 2018 to June 2019. A total of 230 cases and 230 controls participated in the study. Data were analysed using SPSS software. Multivariable logistic model based analysis was conducted to control the effect of potential confounding factors. ORs and 95% CI for the likelihood of developing breast cancer were calculated. RESULTS: The odds of breast cancer was higher among women between 40 and 49 years (adjusted OR (AOR): 3.29, 95% CI 1.39 to 7.77), and being unemployed (AOR: 4.28, 95% CI 2.00 to 9.16). Regarding life style risk factors, women consuming solid oil and using wood or animal dung as source of fuel had significantly higher odds of breast cancer. In addition, the odds of breast cancer was significantly higher among postmenopausal women, women who had previous benign surgery and women with early menarche (<12 years). On the other hand, the odd of breast cancer was significantly lower among women who had moderate physical activities. CONCLUSION: This study showed that occupational status, consumption of solid oil, and using wood or animal dung as source of fuel, early menarche, menopausal status and previous benign breast surgery were associated with breast cancer. On the other hand, physical activity was protective factor. Therefore, there is a need to design appropriate intervention to educate women about life style change or behaviour modification to decrease their breast cancer risk.

Adherence to Chemotherapy among Women with Breast Cancer Treated at Tikur Anbessa Specialized and Teaching Hospital, Addis Ababa, Ethiopia.

BACKGROUND: Adherence is important for women with breast cancer because it is a primary determinant for effectiveness of treatment and optimum clinical benefit. Though Breast cancer is the leading cancer in Ethiopia,adherance to chemotherapy is not investigated in Ethiopian women. OBJECTIVE: This study aimed to assess adherence to chemotherapy among women with breast cancer treated at Tikur Anbessa specialized and Teaching Hospital. METHODS: Cross-sectional study was conducted among 164 breast cancer patients with chemotherapy. After eligible participants were identified, data were collected using face-to-face interviews, card reviews and telephone interviews. Adherence was calculated as the number of doses taken divided by number of recommended or expected doses. Pearson chi-square test was used to evaluate predictors of adherence. RESULTS: Among a total of 164 breast cancer patients, majority, 119, (72.6%) of them were urban residents. The mean age of study participants was 41.99 + 10.9 years. The majority 149, (90.9%) of patients were married. More than half 94, (57.3%) of the women were literate. In this study, 137 out of 164 (83.5%) women were adherent to their chemotherapy. Of the 27 non adherent participants. he reason for non-adherence to chemotherapy was unknown for 7, (25.9%) of women. Among different identified reasons for non-adherent, sever illness prevents patients to receive chemotherapy. Based on Pearson chi square test, distance from referral center and treatment regimen were significantly associated with non-adherence rate. CONCLUSION: The present  study the results showed that the majority 137, (83.5%) of patients were in good adherence to their chemotherapy. The most identified factor of non-adherence was inability to come for their therapy as a result of severity of illness. Therefore, expansion of cancer diagnosis and treatment centers should be encouraged in order to maximize patient's access and adherence to chemotherapy.

Lightweight Multireceptive Field CNN for 12-Lead ECG Signal Classification.

The electrical activity produced during the heartbeat is measured and recorded by an ECG. Cardiologists can interpret the ECG machine's signals and determine the heart's health condition and related causes of ECG signal abnormalities. However, cardiologist shortage is a challenge in both developing and developed countries. Moreover, the experience of a cardiologist matters in the accurate interpretation of the ECG signal, as the interpretation of ECG is quite tricky even for experienced doctors. Therefore, developing computer-aided ECG interpretation is required for its wide-reaching effect. 12-lead ECG generates a 1D signal with 12 channels among the well-known time-series data. Classical machine learning can develop automatic detection, but deep learning is more effective in the classification task. 1D-CNN is being widely used for CVDS detection from ECG datasets. However, adopting a deep learning model designed for computer vision can be problematic because of its massive parameters and the need for many samples to train. In many detection tasks ranging from semantic segmentation of medical images to time-series data classification, multireceptive field CNN has improved performance. Notably, the nature of the ECG dataset made performance improvement possible by using a multireceptive field CNN (MRF-CNN). Using MRF-CNN, it is possible to design a model that considers semantic context information within ECG signals with different sizes. As a result, this study has designed a multireceptive field CNN architecture for ECG classification. The proposed multireceptive field CNN architecture can improve the performance of ECG signal classification. We have achieved a 0.72 F 1 score and 0.93 AUC for 5 superclasses, a 0.46 F 1 score and 0.92 AUC for 20 subclasses, and a 0.31 F 1 score and 0.92 AUC for all the diagnostic classes of the PTB-XL dataset.

Interpretable Machine Learning Techniques in ECG-Based Heart Disease Classification: A Systematic Review.

Heart disease is one of the leading causes of mortality throughout the world. Among the different heart diagnosis techniques, an electrocardiogram (ECG) is the least expensive non-invasive procedure. However, the following are challenges: the scarcity of medical experts, the complexity of ECG interpretations, the manifestation similarities of heart disease in ECG signals, and heart disease comorbidity. Machine learning algorithms are viable alternatives to the traditional diagnoses of heart disease from ECG signals. However, the black box nature of complex machine learning algorithms and the difficulty in explaining a model's outcomes are obstacles for medical practitioners in having confidence in machine learning models. This observation paves the way for interpretable machine learning (IML) models as diagnostic tools that can build a physician's trust and provide evidence-based diagnoses. Therefore, in this systematic literature review, we studied and analyzed the research landscape in interpretable machine learning techniques by focusing on heart disease diagnosis from an ECG signal. In this regard, the contribution of our work is manifold; first, we present an elaborate discussion on interpretable machine learning techniques. In addition, we identify and characterize ECG signal recording datasets that are readily available for machine learning-based tasks. Furthermore, we identify the progress that has been achieved in ECG signal interpretation using IML techniques. Finally, we discuss the limitations and challenges of IML techniques in interpreting ECG signals.

Diagnosis of Diabetes Mellitus Using Gradient Boosting Machine (LightGBM).

Diabetes mellitus (DM) is a severe chronic disease that affects human health and has a high prevalence worldwide. Research has shown that half of the diabetic people throughout the world are unaware that they have DM and its complications are increasing, which presents new research challenges and opportunities. In this paper, we propose a preemptive diagnosis method for diabetes mellitus (DM) to assist or complement the early recognition of the disease in countries with low medical expert densities. Diabetes data are collected from the Zewditu Memorial Hospital (ZMHDD) in Addis Ababa, Ethiopia. Light Gradient Boosting Machine (LightGBM) is one of the most recent successful research findings for the gradient boosting framework that uses tree-based learning algorithms. It has low computational complexity and, therefore, is suited for applications in limited capacity regions such as Ethiopia. Thus, in this study, we apply the principle of LightGBM to develop an accurate model for the diagnosis of diabetes. The experimental results show that the prepared diabetes dataset is informative to predict the condition of diabetes mellitus. With accuracy, AUC, sensitivity, and specificity of 98.1%, 98.1%, 99.9%, and 96.3%, respectively, the LightGBM model outperformed KNN, SVM, NB, Bagging, RF, and XGBoost in the case of the ZMHDD dataset.

A Survey of Brain Tumor Segmentation and Classification Algorithms.

A brain Magnetic resonance imaging (MRI) scan of a single individual consists of several slices across the 3D anatomical view. Therefore, manual segmentation of brain tumors from magnetic resonance (MR) images is a challenging and time-consuming task. In addition, an automated brain tumor classification from an MRI scan is non-invasive so that it avoids biopsy and make the diagnosis process safer. Since the beginning of this millennia and late nineties, the effort of the research community to come-up with automatic brain tumor segmentation and classification method has been tremendous. As a result, there are ample literature on the area focusing on segmentation using region growing, traditional machine learning and deep learning methods. Similarly, a number of tasks have been performed in the area of brain tumor classification into their respective histological type, and an impressive performance results have been obtained. Considering state of-the-art methods and their performance, the purpose of this paper is to provide a comprehensive survey of three, recently proposed, major brain tumor segmentation and classification model techniques, namely, region growing, shallow machine learning and deep learning. The established works included in this survey also covers technical aspects such as the strengths and weaknesses of different approaches, pre- and post-processing techniques, feature extraction, datasets, and models' performance evaluation metrics.

Enhanced Region Growing for Brain Tumor MR Image Segmentation.

A brain tumor is one of the foremost reasons for the rise in mortality among children and adults. A brain tumor is a mass of tissue that propagates out of control of the normal forces that regulate growth inside the brain. A brain tumor appears when one type of cell changes from its normal characteristics and grows and multiplies abnormally. The unusual growth of cells within the brain or inside the skull, which can be cancerous or non-cancerous has been the reason for the death of adults in developed countries and children in under developing countries like Ethiopia. The studies have shown that the region growing algorithm initializes the seed point either manually or semi-manually which as a result affects the segmentation result. However, in this paper, we proposed an enhanced region-growing algorithm for the automatic seed point initialization. The proposed approach's performance was compared with the state-of-the-art deep learning algorithms using the common dataset, BRATS2015. In the proposed approach, we applied a thresholding technique to strip the skull from each input brain image. After the skull is stripped the brain image is divided into 8 blocks. Then, for each block, we computed the mean intensities and from which the five blocks with maximum mean intensities were selected out of the eight blocks. Next, the five maximum mean intensities were used as a seed point for the region growing algorithm separately and obtained five different regions of interest (ROIs) for each skull stripped input brain image. The five ROIs generated using the proposed approach were evaluated using dice similarity score (DSS), intersection over union (IoU), and accuracy (Acc) against the ground truth (GT), and the best region of interest is selected as a final ROI. Finally, the final ROI was compared with different state-of-the-art deep learning algorithms and region-based segmentation algorithms in terms of DSS. Our proposed approach was validated in three different experimental setups. In the first experimental setup where 15 randomly selected brain images were used for testing and achieved a DSS value of 0.89. In the second and third experimental setups, the proposed approach scored a DSS value of 0.90 and 0.80 for 12 randomly selected and 800 brain images respectively. The average DSS value for the three experimental setups was 0.86.

Deep Learning in Selected Cancers' Image Analysis-A Survey.

Deep learning algorithms have become the first choice as an approach to medical image analysis, face recognition, and emotion recognition. In this survey, several deep-learning-based approaches applied to breast cancer, cervical cancer, brain tumor, colon and lung cancers are studied and reviewed. Deep learning has been applied in almost all of the imaging modalities used for cervical and breast cancers and MRIs for the brain tumor. The result of the review process indicated that deep learning methods have achieved state-of-the-art in tumor detection, segmentation, feature extraction and classification. As presented in this paper, the deep learning approaches were used in three different modes that include training from scratch, transfer learning through freezing some layers of the deep learning network and modifying the architecture to reduce the number of parameters existing in the network. Moreover, the application of deep learning to imaging devices for the detection of various cancer cases has been studied by researchers affiliated to academic and medical institutes in economically developed countries; while, the study has not had much attention in Africa despite the dramatic soar of cancer risks in the continent.

Quality of life and associated factors among patients with breast cancer under chemotherapy at Tikur Anbessa specialized hospital, Addis Ababa, Ethiopia.

BACKGROUND: Breast cancer is the most common cancer affecting women in Ethiopia with increasing burden, and chemotherapy treatment produces a detrimental effect on individual wellbeing. Since last few years quality of life has been the primary goal of cancer treatment, yet little research has been conducted on quality of life of breast cancer patients under chemotherapy. OBJECTIVE: To determine the quality of life and associated factors among patients with breast cancer under chemotherapy at Tikur Anbessa specialized hospital, Addis Ababa, Ethiopia. METHODS: Institution based cross-sectional study was conducted on 404 patients with breast cancer, who took at least one cycle of chemotherapy treatment using face to face interview at oncology unit of Tikur Anbessa specialized hospital day care center from February to April 2018. The validated Amharic version of European organization for research and treatment of cancer core 30 (EORTC QLQ-C30) and quality of life questionnaire specific to breast (QLQ-BR23) was used to measure health related quality of life. Both descriptive and inferential statistics were used. For the purpose of interpretation quality of life score was dichotomized in to two using the calculated mean score, which is 53 as a cutoff point, then, bi-variable and multivariable logistic regression was used to describe association between dependent and independent variables. Hence, patients who score above 53 for quality of life were considered to have good quality of life. RESULT: Of the total sample, overall response rate was 99.77%. The average quality of life score of patients with breast cancer under chemotherapy treatment was 52.98 (SD = 25.61). Majority of patients had scored poor in emotional functioning, sexual functioning, and financial difficulties. Educational status of college and above, being divorced, higher household income, higher scores of physical and social functioning were associated with significantly improved (better) quality of life. Lower scores of fatigue, insomnia, financial difficulties and systemic therapy side effects all were associated with better scores of quality of life of breast cancer patients. Whereas, patients receiving < = 2 cycles of chemotherapy had significantly lower scores of quality of life. CONCLUSION AND RECOMMENDATION: Quality of life of breast cancer patients under chemotherapy treatment is poor in comparison with the reference data and international findings. Therefore, quality of life assessment should be incorporated in patient's treatment protocol. And financial aids may significantly improve the quality of life of breast cancer patients under chemotherapy treatment.