iosr-JCE   Volume-24 ~ Issue-2 ~ Mar. - Apr. 2022

Abstract: Fetal heart disease is the most common and serious congenital anomaly in newborns, and the incidence rate is about 0.4%~1.3% of newborns. In recent years, the incidence rate of shows a significant increasing trend.It is one of the leading causes of death in newborns and in childhood.Traditional manual delineation of cardiac occupation is not only time-consuming, but also not accurate enough.However, current deep learning algorithms rely mainly on training data and require large amounts of manually labeled data for training.Due to the large amount of medical image data, manual labeling is very time-consuming and laborious, and subjective errors may occur, which limits the application of deep learning methods in the field of medical image segmentation.Therefore, this paper presents an automatic segmentation method of echo-cardiogram images of the fetal heart based on a semi-supervised machine learning method.Compared to state-of-theart, our method shows cardiac space segmentation can be processed quickly, so as to evaluate cardiac function and provide an important guarantee for eugenics.

Keywords— Fetal heart, Echo-cardiogram, image segmentation, semi-supervised machine learning

[1]. Hanlon-Dearman A , Green C R , Andrew G , et al. Anticipatory Guidance for Children and Adolescents with Fetal Alcohol Spectrum Disorder (FASD): Practice Points for Primary Health Care Providers[J]. The Canadian journal of clinical pharmacology = Journal canadien de pharmacologie clinique, 2015, 22(1):e27-56.
[2]. Tsysar S A , Nikolaev D A , Sapozhnikov O A . Broadband Vibrometry of a Two-Dimensional Ultrasound Array Using Transient Acoustic Holography[J]. Acoustical Physics, 2021, 67(3):320-328.
[3]. Roodt E . Job demands and job resources as predictors of dispositional employability of academics in South Africa[J]. European Heart Journal Cardiovascular Imaging, 2013, 14(2):142-148.
[4]. Mudaliar A V , Kashyap R S , Purohit H J , et al. Detection of 65 kD heat shock protein in cerebrospinal fluid of tuberculous meningitis patients[J]. BMC Neurology, 2006, 6.
[5]. Elad M. Sparse and redundant representation modeling- what next? IEEE Signal Process. Lett., 2012; 19(12): 922-928.

Abstract:The proposed system's system design and procedural details are described in this paper. The preliminary and comprehensive design of the system is used to define its workflow. The thorough design explains how the system's components and sub-components interact. The system's procedural details define the whole working procedure, from data collection through analysis and estimation, in step-by-step order. Background: This paper delves into the K-Means and K-Prototype clustering algorithms for consumer behavioural and segmentation models. It comprises the segmentation's theoretical foundation as well as existing E-SAT-based systems. It allows for the statistical analysis of their data properties to be compared. The study went into great detail into the theoretical and algorithmic features of recent customer behavioural and segmentation approaches for E-satisfaction, particularly employing K-Prototype clustering. It depicts the path taken by such systems.....

Key Word: Customer Behavior; K-Prototype; Satisfaction, Customer Segmentation; Machine Learning.

[1]. C. J. Udokwu, F. Darbanian, N. F. M. Falatouri, and P. Brandtner. Evaluating technique for capturing customer satisfaction data in retail supply chain. The 4th International Conference on E-commerce, E-Business and E-Government, ACM, NY, USA, pp. 89–95, 2020.
[2]. H-F. Chen, and S-H Chen. "How Website Quality, Service Quality, Perceived Risk and Customer Satisfaction Affects Repurchase Intension? A Case of Taobao Online Shopping", In: Proceedings of the 10th International Conference on E-Education, E-Business, E-Management and E-Learning, ACM, NY, USA, 326–328, 2019.
[3]. S. Zhang, X. Tan, J. Wang, J. Chen, and X. Lai. "Modeling Customers' Loyalty using Ten Years' Automobile Repair and Maintenance Data: Machine Learning Approaches", In: Proceedings of the 2019 2nd International Conference on Data Science and Information Technology, ACM, NY, USA, pp. 242–248, 2019.
[4]. M. Zhao, Q. Zeng, M. Chang, Q. Tong, and J. Su, "A Prediction Model of Customer Churn Considering Customer Value: An Empirical Research of Telecom Industry in China", Discrete Dynamics in Nature and Society, Vol. 2021, pp. 1-12, 2021.
[5]. A. Ahmad, and O. Barukab. "A Novel Approach to Identify the Categories of Attributes for the Three-Factor Structure in Customer Satisfaction", Complexity, Vol. 2020, pp. 1-12, 2020..

Abstract: Data Analysis is the process of bringing order and structure to collected data. It turns data into information teams can use. Data visualization is the process of putting data into a chart, graph, or other visual format that helps inform analysis and interpretation. Analysis and Visualization of datasets has always been a helpful for various reasons whether it's for improvement of customer experience or business plans, etc. These all aspects require the analysis of the data......

Key Word: Data Analysis, NumPy, Data Visualization, Pandas, Dataset.

[1]. Front. Public Health, 28 May 2020, https://doi.org/10.3389/fpubh.2020.00216
[2]. Cyranoski D, "Did pangolins spread the China coronavirus to people?" - Nature. (2020), https://www.nature.com/articles/d41586-020-00364-2
[3]. "COVID-19: Emergence, Spread, Possible Treatments, and Global Burden", https://www.frontiersin.org/articles/10.3389/fpubh.2020.00216/full#B13
[4]. https://covid19.who.int
[5]. https://www.mohfw.gov.in

Abstract: Object detection is a computer vision method that locates the position of the object in the image frame. Deep Learning methods have recently made a remarkable breakthrough in many issues such as computer vision, speech recognition and natural language processing. In Deep Learning, Convolution Neural Network is a multi-layer neural network that helps us in the Object Detection process. We trained a model that detects the different surface regions using the Convolution Neural Network with high accuracy. This work is carried out for totally and partially visually impaired people in order to help them classify the path they are walking on. The model used detects the surface regions and notifies the sightless person through text to sound conversion API. This adds vision to the visionless person. In this paper, we have developed a model that detects different path texture for visually impaired people using Faster RCNN, RESNET and Inception Object Detection models from the given input and predicts which model will achieve greater accuracy. The least loss we obtained was 0.02 from Faster RCNN.

Key Word: Object detection, Path classification, Visually Impaired, RCNN, RESNET, Computer Vision, Deep Learning, Dataset.

[1]. World Health Organization, GLOBAL DATA ON VISUAL IMPAIRMENTS : Fact sheet number 3, 2010. [Online]. https://www.who.int/blindness/GLOBALDATAFINALforweb.pdf?ua=1
[2]. Blasch, B.B., Wiener, W.R., & Welsh, R.L., " Foundations of orientation and mobility," (2nd ed.). NY: AFB Press, New York, 1997.
[3]. Marion, A.H., Michael, A.J., "Assistive technology for Visuallyimpaired and Blind People," Springer, London, UK, 2008.
[4]. J. Bai, S. Lian, Z. Liu, K. Wang and D. Liu, "Smart guiding glasses for visually impaired people in indoor environment," in IEEE Transactions on Consumer Electronics, vol. 63, no. 3, pp. 258-266, August 2017, doi: 10.1109/TCE.2017.014980.
[5]. Kaushalya, V. S. S., K. Premarathne, H. M. Shadir, P. R. Krithika and S. G. S. Fernando. "" AKSHI " : Automated Help aid for Visually Impaired People using Obstacle Detection and GPS Technology." (2016).

Abstract: Sentiment Analysis or opinion mining in general, is the use of natural language processing, computer linguistics and related technology used to determine whether the text data is positive, neutral or negative. Hate Speech, on the other hand is abusive or threatening speech or writing that expresses prejudice, hate and / or encourages violence towards a person or a group of persons based on their race, religion, sex, or sexual orientation. With the exponential rise in the number of people making use of social media, where tons of content is posted daily, which is visibly harmless in nature, there has been a sharp rise in hate speech as well. The need and interest for identifying and detecting hate speech on such social media platforms, especially Twitter has risen significantly, as now major firms move towards development of systems which help tackle hate speech online. This project aims to cater to the need of identifying negative tweets which promote hate speech.

[1]. Paul Rottger, Bertram Vidgen, Dong Nguyen, Zeerak Waseem,Helen Margetts, Janet B. Pierrehumbert, Uni-versity of Oxford, The Alan Turing Institute, Utrecht University, University of Sheffield: HATECHECK: Functional Tests for Hate Speech Detection Models, May 2021.
[2]. Soroosh Tayebi Arasteh, Mehrpad Monajem, Vin-cent Christlein, Philipp Heinrich, Anguelos Nicolaou, Hamidreza Naderi Boldaji, Mahshad Lotfinia and Ste-fan Evert, Friedrich-Alexander-Universitat Erlangen-Nurnberg, Germany, Harvard Medical School, United States, Sharif University of Technology, Iran: How will your Tweets be Recieved? April 2021.
[3]. Abdullah Alsaeedi, Mohammad Zubair Khan, A Study on Sentiment Analysis Techniques of Twitter Data - International Journal of Advanced Computer Science and Applications (IJACSA) Volume No. 2, November 2019.
[4]. Ona de Gibert, Naiara Perez, Aitor Garc´ıa-Pablos, Montse Cuadros, Hate Speech Dataset from a White Supremacy Forum HSLT Group at Vicomtech, Donos-tia/San Sebasti´an, Spain, September 2018.
[5]. Pedro H. Calais, Yuri A. Santos, Virg´ılio A. F. Almeida, Wagner Meira Jr., Universidade Federal de Minas Gerais Belo Horizonte, Minas Gerais, Brazil, Characterizing and Detecting Hateful Users on Twitter - Manoel Horta Ribeiro, March 2018.

Abstract: Background: The RADAR MIMO uses multiple antennas on emitter and receiver for avoiding doppler effects, improving the robustness of binary error of transmission. Multi carrier modulation is introduced to have good performance about doppler effects like Universal Filtered Multicarrier (UFMC) used specially on 5G. This modulation is like Orthogonal Frequency Division Multiplex (OFDM) but divides into multiple sub-bands. The problem of OFDM and UFMC is concerned about Discret Fourier Transform (DFT) which gives Peak Average Power Ratio (PAPR) and high time processing. This article replaces the DFT by Discret Cosine Transform (DCT) and Lift Wavelet......

Key Word: RADAR; MIMO; LWT; UFMC; PAPR..

[1]. Zhou, Zhenlei, Lin, Bangjiang, Tang, Xuan, Chaudhary, Sushank, Lin, Chun, "Performance comparison of DFT-OFDM, DCT-OFDM, and DWT-OFDM for visible light communications", spie. Journal, International Conference on Optical Communications and Networks, 2018, Zuhai China.
[2]. Javaid A. Sheikh, Farhana Mustafa, Arshid Iqbal "Resource Allocation of Power in FBMC based 5G Networks using Fuzzy Rule Base System and Wavelet Transform", International Journal of Advanced Research in Science and Engineering (IJARSE), march 2018.
[3]. Zainab Hdeib Al-Shably, Zahir M. Hussain, "Performance of FFT-OFDM versus DWT-OFDM under Compressive Sensing", Journal of Physics: Conference Series, 2021
[4]. Khaizuran Abdullah, Zahir M. Hussain, "Studies on DWT-OFDM and FFT-OFDM Systems", International Conference on Communication, Computer and Power (ICCCP 2009)
[5]. Elif Büşra Tuna, Yusuf İslam Tek, Ali Ozen, "A Novel Approach based on Lifting Wavelet Transform for MIMO-OFDM Systems", Research square 2021.