Prediction of Students’ Performance based on Academic, Behaviour, Extra and Co-Curricular Activities

Since Academic institutions contain huge volume of data regarding students such as academic scores, scores in co and extracurricular activities, family annual income, family background and other supporting documents, predicting individual students performance in all aspects manually is a difficult task. The proposed work uses data mining techniques to identify students who are eligible for scholarships and other benefits. Students are classified into different categories by means of academic, behavior, extra and co-curricular activities. Machine Learning algorithms such as Naive Bayes, Decision Tree Classifier and Support Vector Machine are used for predicting the performance of the student. With the help of this proposed model parents and instructors can monitor student’s performance and they can also provide essential technical and moral support. Also this helps in providing academic scholarship and training to the students to support them financially and to enrich their knowledge. It suggests the Academic Institutions to organize induction or training programmes at the beginning of the semester. Technical training, motivational talks, Yoga, etc are organized by the institutions by keeping in mind of students physical and mental health. Considering the e-learning platforms huge volumes of data and plethora of information are generated. In this work, various learning models are constructed and their accuracies are compared to analyse which algorithm out-performs.

Effectiveness of neurobic exercise program on memory and depression among elderly residing at old age home

Objectives This study finds out the effectiveness of neurobic exercise program on memory and depression among elderly residing in old age homes. Methods The non-probability purposive sampling technique was used for sample selection. Wechsler's memory scale (WMS-IV) and Geriatric depression scale (GDS) were the instruments used to assess the memory and depression among elderly during the pretest and posttest, respectively and the researcher had developed data sheet to collect information about the background variables using interview technique. Results The neurobic exercise program was found to be effective in reducing depression among elderly residing in old age homes. There was a significant difference (p<0.001) in the level of depression had been found during the pretest and posttest in the interventional group. There was a statistically significant difference (p<0.001) found between the study group and in the control group. There was significant correlation (r=0.417, p<0.05) found between the memory and depression during the pretest in the study group among the elderly. A statistically significant association (p<0.05) found in the mean scores of depression and marital status of the elderly during the pretest in the study group and there was a significant association (p<0.01) found in the mean scores of depression and the gender of the elderly during the pretest and posttest in the non interventional group were found. Conclusions The findings suggested that neurobic exercise program is an effective intervention in improving memory and reducing depression.

An Investigation of Bayes Algorithm and Neural Networks for Identifying the Breast Cancer

Context

Breast cancer is a biggest threat to women. X-ray mammography is the most effective method for early detection and screening of breast cancer. It is a tough challenge for the radiologist in reading mammography since it does not provide consistent result every time.

Aim

To improve the primary sign of this disease, computer-aided diagnosis schemes have been developed. Using monitor, digital images of mammography are displayed and they can be lightened or darkened before they are printed on the film. Time factor is important to identify the abnormality in body such as breast cancer and lung cancer. Hence, to detect the tissues and treatment stages, image-processing techniques are improved in several medical areas. In this project, using low-level preprocessing techniques and image segmentation, the breast cancer detection is done.

Methods

With the help of Bayes algorithm and neural networks (NNs), the type of the mammogram and stages is identified. For segmentation process, region-growing algorithm is used, which helps to find the affected portion, i.e., region of interest. Gray-level co-occurrence matrix (GLCM) and texture feature are used for feature extraction.

Results

Bayes algorithm is used for probability of identification, whereas NNs is used to reduce the probability level from 0-1000 to 0-1 in case of classification.

Design and development of compact monitoring system for disaster remote health centres

PURPOSE

To enhance speedy communication between the patient and the doctor through newly proposed routing protocol at the mobile node.

MATERIALS AND METHODS

The proposed model is applied for a telemedicine application during disaster recovery management. In this paper, Energy Efficient Link Stability Routing Protocol (EELSRP) has been developed by simulation and real time. This framework is designed for the immediate healing of affected persons in remote areas, especially at the time of the disaster where there is no hospital proximity. In case of disasters, there might be an outbreak of infectious diseases. In such cases, the patient's medical record is also transferred by the field operator from disaster place to the hospital to facilitate the identification of the disease-causing agent and to prescribe the necessary medication. The heterogeneous networking framework provides reliable, energy efficientand speedy communication between the patient and the doctor using the proposed routing protocol at the mobile node.

RESULTS

The performance of the simulation and real time versions of the Energy Efficient Link Stability Routing Protocol (EELSRP) protocol has been analyzed. Experimental results prove the efficiency of the real-time version of EESLRP protocol.

CONCLUSION

The packet delivery ratio and throughput of the real time version of EELSRP protocol is increased by 3% and 10%, respectively, when compared to the simulated version of EELSRP. The end-to-end delay and energy consumption are reduced by 10% and 2% in the real time version of EELSRP.

Cystatin C: An alternative dialysis adequacy marker in high flux hemodialysis

The conventional, low flux (LF) dialyzer allows the removal of small molecular solutes like urea and creatinine. High flux (HF) dialyzers allow the effective removal of middle molecules (MM) as well, and are associated with reduced hemodialysis-related morbidity and mortality. Cystatin C has attractive characteristics as a representative MM. The aim of this study was to determine cystatin C reduction ratio (CysCRR) in both LF and HF groups and to compare it with other markers of dialysis adequacy. Thirty-seven patients were subjected to both LF and HF hemodialysis 2 weeks apart. Serum urea, creatinine and cystatin C were measured pre- and post-dialysis. Cystatin C was measured by latex-enhanced immunoturbidimetry. Urea and creatinine reduction ratios were 72.3 ± 14.7% and 62.5 ± 13%, respectively in the LF group. The CysCRR was −9.7 ± 6.7% and 29.2 ± 11% in LF and HF hemodialysis, respectively. The statistically significant decrease in CysCRR in the HF group shows the effective clearance of MM by HF dialyzers. Hence, CysCRR could be applied to measure the MM clearance in HF hemodialysis. This study highlights the significance of cystatin C as an important dialysis adequacy marker replacing the conventional markers such as urea and creatinine in HF hemodialysis. Among the middle molecules cystatin C scores over beta-2 microglobulin.

Hereditary breast/ovarian cancer: clinicopathological characteristics and survival of BRCA2 positive and negative cases

The clinical and pathological characteristics and prognostic outcome of patients with hereditary breast/ovarian cancer and BRCA2 mutations are poorly known. Hence, the present study aimed to correlate the BRCA2 mutation status with clinical characteristics and overall survival of 102 breast/ovarian cancer patients in Kerala, South India. All the coding regions of BRCA2 genes were PCR amplified and analyzed for mutations employing Conformation Sensitive Gel Electrophoresis and characterized by sequencing. The ORs with 95% Cls was computed to assess the association between BRCA2 gene mutation status and clinicopathologic characteristics of breast cancer patients. Survival curves were generated according to Kaplan-Meier method using Log Rank test and Cox proportional hazards regression method. Out of the 102 breast/ovarian cancer patients with known BRCA2 status, 19 were BRCA2 mutation positive. In survival analysis, BRCA2 gene mutation status (P = 0.02) and clinicopathologic parameters such as tumour size (p = 0.01), metastasis (P = 0.01), disease stage (P = 0.03) and laterality (P = 0.02) were significantly associated with poor prognosis of breast cancer patients. Patients with hereditary breast/ovarian cancer resulting from a BRCA2 mutation have been conclusively shown to have a worse survival prognosis compared to the non mutated group of patients.