Informatics Publishing Limited

K. R. Ananthapadmanabhan ¹, G. Parthiban ¹

Affiliations
1 SRM Arts and Science College, Kattankulathur, Tamilnadu, IN

Abstract

Diabetic retinopathy the most common diabetic eye disease, is caused by complications that occurs when blood vessels in the retina weakens or distracted. It results in loss of vision if early detection is not done. Several data mining technique serves different purposes depending on the modeling objective. The outcome of the various data mining classification techniques was compared using rapid miner tool. We have used Naive bayes and Support Vector Machine to predict the early detection of eye disease diabetic retinopathy and found that Naive bayes method to be 83.37% accurate. The performance was also measured by sensitivity and specificity. The above methodology has also shown that our data mining helps to retrieve useful correlation even from attributes which are not direct indicators of the class which we are trying to predict.

Keywords

Data Mining, Diabetes, Naive Bayes Method, Retinopathy, Support Vector Machine

Full Text

   

T. M. Balakrishnan Nair ¹, V. Ramaswamy ¹, G. Parthiban ¹, R. Shankar ²

Affiliations
1 National Institute of Oceanography, Dona Paula, Goa - 403 004, IN
2 Department of Marine Geology, Mangalore University, Mangalagangotri - 574 199, IN

Abstract

Particle fluxes were measured using six time-series sediment traps at three sites in the western (16°20' N; 60°30' E), central (14°31' N; 64°46' E) and eastern (15°31' N; 68°43' E) Arabian Sea. Trap deployment depths were between 900 and 3000 m and collection period was from December 1992 to February 1994.

Annual particle fluxes showed an east-west trend with minimum fluxes (22.25 gm^-2) in the eastern Arabian Sea and maximum fluxes (69.81 g m^-2) in the western Arabian Sea. Carbonates, contributed mainly by foraminifers and coccolithophorids, are the dominant component in all the traps. Opal fluxes were maximum in the western Arabian Sea. At all the locations, lithogenic percentages increased with depth whereas organic carbon percentages decreased. Particle flux patterns show a strong seasonality with peak fluxes during the southwest (SW) monsoon (June to September). Relatively high fluxes were also observed during the northeast (NE) monsoon (December to February).

In the western Arabian Sea, particle fluxes are dominated mainly by carbonates during the early SW monsoon but by biogenic silica during the fate SW monsoon. The increase in particle fluxes during the early SW monsoon is related to variations in the mixed layer depth which, in turn, is controlled by the strength of the Findtater Jet and the curl of the wind stress. The increase in biogenic silica fluxes during the late SW monsoon is related to the advection of nutrient-rich water from the Oman and Somali upwelling areas. In the eastern Arabian Sea, particle fluxes are high during the NE monsoon due to the effects of winter cooling.

Keywords

Oceanography, Particle Flux, Sediment Trap, Sea Surface Temperature, Arabian Sea.

Full Text

     

J. N. Pattan ¹, G. Parthiban ¹

Affiliations
1 National Institute of Oceanography, Dona Paula, Goa - 403 004, IN

Abstract

Five buried ferromanganese nodules were recovered at different depths of 167-169 cm (A), 172- 174 cm (B), 228-230 cm (C), 328-330 cm (D) and 418-420 cm (E) in a 5 m sediment core from the siticeous sediment of Central Indian Basin (Lat 9°00' S , Long 76°00' E) at a water depth of 5260 m. These nodules were analysed for major, trace and rare earth element(REE) concentrations to understand their genesis and compare with the surface and burled nodules in the top 1 in of the sediment column from the same basin. Nodules A, B, C are of an early diagenetic(Mn/Fe ratio between 9 3 and 15 I), nodule D IS of hydrogenetic(Mn/Fe ratio 1 6) and nodule E is of diagenetic(Mn/Fe ratio 3 0) origin. Total REE concentration ranges between 164 and 497 ppm (av-348 ppm) and is nearly 2 to 3 fold lower than the surface and burled nodules REE are conveyed from the seawater to the nodules In association with a combined phase consistnig of Fe - Ti - P. The shale (NASC) - Normalized REE pattern displays a small convex pattern with a positive Ce-anomaly indicating an oxidized environment while, the presence of positive E u-anomaly in nodules A, B and C probably suggest an aeolian dust. These burled nodules show a moderate middle and heavy REE enrichment (strong in nodules B and C) compared to light REE, but fractionation between middle and heavy REE is not very clear It appears that not much significant Post -Depositional changes have occurred in these buried nodules.

Keywords

Buned Fe-Mn Nodules, Chemical Composlition, Genesis, REE Patterns, Central Indian Ocean Basin.

Full Text

     

G. Parthiban ¹, S. K. Srivatsa ²

Affiliations
1 Dr. MGR Educational Research and Institute, MGR University, Chennai, IN
2 Dept. of E & I, Prathyusha Institute of Technology and Management, Chennai, IN

Abstract

Classifying data is a common task in Machine learning. Data mining in health care is an emerging field of high importance for providing prognosis and a deeper understanding of medical data. Most data mining methods depend on a set of features that define the behaviour of the learning algorithm and directly or indirectly influence the complexity of resulting models. Diabetic retinopathy the most common diabetic eye disease, is caused by complications that occurs when blood vessels in the retina weakens or distracted. We have applied machine learning methods to predict the early detection of eye disease diabetic retinopathy and found that Decision Tree method to be 90% accurate. The performance was also measured by sensitivity and specificity.

Keywords

Data Mining, Naive Bayes Method, Decision Tree, Diabetes, Diabetic Retinopathy.

Full Text

     

D. Magdalene Delighta Angeline ¹, G. Parthiban ¹, K. Utchimakali ¹, V. Gideon Augustin ¹

Affiliations
1 Department of Computer Science and Engineering, Dr.G.U. Pope College of Engineering, Sawyerpuram-628251, Tamilnadu, IN

Abstract

Face emotion is required in many applications like eye-gaze tracking, iris detection, video conferencing, auto-stereoscopic displays, face detection and face recognition. This paper proposes a novel technique for eye detection using color and morphological image processing. It is observed that eye regions in an image are characterized by low illumination, high density edges and high contrast as compared to other parts of the face. The method proposed is based on assumption that a frontal face image (full frontal) is available. Firstly, the skin region is detected using a color based training algorithm and six-sigma technique operated on RGB, HSV and NTSC scales. Further analysis involves morphological processing using boundary region detection and detection of light source reflection by an eye, commonly known as an eye dot. This gives a finite number of eye candidates from which noise is subsequently removed. This technique is found to be highly efficient and accurate for detecting eyes in frontal face images.

Keywords

Colour, Content Based Image Retrieval, Gray Scale, Phong Shading, Texture.

Full Text

     

K. R. Anantha Padmanaban ¹, G. Parthiban ¹

Affiliations
1 Department of Computer Science and Applications, SRM Arts and Science College, Kattankulathur - 603203, Tamil Nadu, IN

Abstract

Objective: This paper aims at predicting the early detection of chronic kidney disease also known as chronic renal disease for diabetic patients with the help of machine learning methods and finally suggests a decision tree to arrive at concrete results with desirable accuracy by measuring its performance to its specification and sensitiveness. Methods: The behaviour of learning algorithms determined on a set of data mining indicators has a proportionate effect on the resulting models. Discovering the knowledge from wide databases is termed as Data mining. Besides studying the existing available Clinic Foundation Heart Disease dataset, 600 clinical records collected by us from a leading Chennai based diabetes research centre. We have tested the dataset for classification using Naïve Bayes and Decision tree method. Findings: On comparing the classification algorithms with respect to Naïve Bayes and Decision tree, we came to conclusion that the accuracy is up to 91% for Decision tree classification. Applications/Improvement: In order to increase the accuracy of the prediction result, we have utilized algorithms such as neural network and clustering data which greatly helped in our mission and also gave scope for future research.

Keywords

Chronic Kidney Disease (CKD), Data Mining, Decision Tree, Diabetes, Naïve Bayes Method.

Full Text