Kanakaprabha S

Kanakaprabha S

Alappuzha, Kerala, India
452 followers 390 connections

Join to follow

About

As a passionate Data Scientist and AI Enthusiast, I am deeply committed to leveraging Machine Learning and Deep Learning techniques to solve complex problems. My journey in AI has been driven by a profound interest in applying these technologies to medical imaging, with a special focus on advancing diagnostic accuracy and early disease detection. My academic background and hands-on experience have equipped me with a robust foundation in AI research, and I am eager to contribute to innovative projects that push the boundaries of healthcare technology. Driven by curiosity and a strong desire to learn, I am enthusiastic about exploring new challenges and opportunities in the field.

Articles by Kanakaprabha

Activity

Experience

Education

  • Graphic

    -

  • -

Licenses & Certifications

Publications

  • Parkinson Disease Detection Using Various Machine Learning Algorithms

    IEEE

    Parkinson disease is a neural disease. It prompts shaking of the hands, difficulty to walk, balance with coordination. No medical treatment is available in the high-level stage. X-ray, CT scan and blood tests report are not sufficiently results available in the early stage. About two trillion community are alive in Parkinson's disease (PD) in the U.K., which is the highest number of people affected. are pinpointed to have different sclerosis, solid dystrophy and Lou Gehrig's illness. This is…

    Parkinson disease is a neural disease. It prompts shaking of the hands, difficulty to walk, balance with coordination. No medical treatment is available in the high-level stage. X-ray, CT scan and blood tests report are not sufficiently results available in the early stage. About two trillion community are alive in Parkinson's disease (PD) in the U.K., which is the highest number of people affected. are pinpointed to have different sclerosis, solid dystrophy and Lou Gehrig's illness. This is relied upon to ascend to 1.5 million by 2040. Around the 75,000 Americans are diagnosis PD with every year. It is very important to predict Parkinson's disease early so that important treatment can be done. The purpose of the proposed work is to detect Parkinson disease, where we aimed to identify disease in early prediction using clinical imaging that incorporate the use of Machine learning techniques. A comparative analysis done with various Machine Learning classifier algorithms like XGBoost, Random Forest, KNN, SVM are the best model is proposed which is used to make predictions and find accuracy. We are observed that Random Forest provides better performance with an accuracy 90%. Automatic detection with more accuracy will make screening for Parkinson disease as cost effective and efficient manner facilitates to use appropriate and fast solutions.

    See publication

  • Analysis of COVID-19 and Pneumonia Detection in Chest X-Ray Images using Deep Learning

    IEEE

    The on-going COVID-19 outbreak made healthcare systems across the globe to be in the edge of the battle. Recent stats indicate that more than 140 million confirmed cases are diagnosed globally as of April 2021. The cases are increasing day by day. The early and auto diagnosis helps people to be precautious. The proposed work aims to detect COVID-19 patients and Pneumonia patients from X-Rays which is one of the medical imaging modes to analyse the health of patient’s lung inflammation. The…

    The on-going COVID-19 outbreak made healthcare systems across the globe to be in the edge of the battle. Recent stats indicate that more than 140 million confirmed cases are diagnosed globally as of April 2021. The cases are increasing day by day. The early and auto diagnosis helps people to be precautious. The proposed work aims to detect COVID-19 patients and Pneumonia patients from X-Rays which is one of the medical imaging modes to analyse the health of patient’s lung inflammation. The suitable Convolutional Neural Network Model is selected for the identified dataset. The model detects COVID-19 patients and Pneumonia patients on the real-world dataset of lung X-Ray images. Images are pre-processed and trained for various classifications like Normal, COVID-19 and Pneumonia. After pre-processing, the detection of the disease is done by selecting the appropriate features from the images in each of the datasets. The result indicates that accuracy of detection of COVID vs Normal and COVID vs Pneumonia. Among those two, COVID vs Normal is with better accuracy than COVID vs Pneumonia. This method detects not only COVID or Pneumonia, but also the subtypes of Pneumonia as bacterial or Viral Pneumonia with 80% and 91.46% respectively. The detection of COVID, Bacterial Pneumonia and Viral Pneumonia using the proposed model helps in rapid diagnosis and to distinguish COVID from Pneumonia and its types which facilitates to use appropriate and fast solutions.

    See publication

  • Journal

    International Journal of Engineering Researchers and Management Studiesmentnd

    See publication

Patents

  • INNOVATIVE CYBER-PHYSICAL SYSTEM BASED ON THE INTERNET OF THINGS (IOT) AND CLOUD COMPUTING TO MONITOR WATER QUALITY IN RURAL AREAS

    Issued 202321017111

    As a result of the method in which markets operate being altered by new
    technologies, there is greater pressure on the sector to become more flexible
    and adaptable in order to meet the markets' ever-changing needs. Businesses
    are undergoing considerable changes to their production methods, technology
    infrastructures, and management styles in order to meet the increasing
    efficiency, productivity, and quality requirements of the global market.
    Automation, digitalization…

    As a result of the method in which markets operate being altered by new
    technologies, there is greater pressure on the sector to become more flexible
    and adaptable in order to meet the markets' ever-changing needs. Businesses
    are undergoing considerable changes to their production methods, technology
    infrastructures, and management styles in order to meet the increasing
    efficiency, productivity, and quality requirements of the global market.
    Automation, digitalization, and artificial intelligence, along with the widespread
    availability and low cost of computing power, smart sensors, data acquisition
    systems, intelligent robotics, information and communication technology, the
    Internet of Things, Big Data, and cloud computing, have all contributed to
    enhanced performance and productivity across all industries. The majority of
    efficiency and output gains may be attributed to technology advances in these
    areas. Intelligent manufacturing is replacing conventional production as a
    result of digitization and other advances in cutting-edge technology. As a result
    of these changes, we are currently seeing "Industry 4.0," or the fourth
    industrial revolution. Humans in the modern world require increasing
    quantities of potable water. In contrast, climate change is destabilising the
    water cycle, which is damaging to water supply. Climate change is responsible
    for this disaster. As a result, worldwide water scarcity has become a major
    concern. As the global water crisis deepens, water utilities must adopt new
    technical paradigms to maximise the use of the available water. The ultimate
    goal of the construction of huge, complex water delivery systems, also known
    as water supply networks, by water utilities across the nation is to provide
    clean drinking water to all citizens. According to the United Nations'
    Sustainable Development Goals, obtaining clean water is an important
    objective

    See patent

  • IMPLEMENTING MACHINE LEARNING AND DEEP NEURAL NETWORKS TO ANALYSE STOCK MARKET TRENDS AND ASSESS THE ACCURACY OF MARKET PREDICTIONS

    Issued 202341001183

    See patent

  • ARTIFICIAL INTELLIGENCE AND CLOUD BASED SMART HR MANAGEMENT SYSTEM IN FINANCIAL SECTOR USING MACHINE LEARNING ALGORITHMS

    Issued 202221062006

    Cloud computing, machine learning, and artificial intelligence innovations are
    reshaping many facets of business, such as finance, healthcare, human
    resources, marketing, etc. People have suggested that technologies like AI, ML,
    and cloud have the potential to revolutionise the sport. Banking, healthcare,
    human resources, and marketing will be among the first industries impacted
    by artificial intelligence. AI and its enabling technologies, such as machine
    learning and…

    Cloud computing, machine learning, and artificial intelligence innovations are
    reshaping many facets of business, such as finance, healthcare, human
    resources, marketing, etc. People have suggested that technologies like AI, ML,
    and cloud have the potential to revolutionise the sport. Banking, healthcare,
    human resources, and marketing will be among the first industries impacted
    by artificial intelligence. AI and its enabling technologies, such as machine
    learning and deep learning, have ushered in a new era of intelligent automation
    and human-level recognition. AI is already disrupting industries such as
    finance, healthcare, human resources (HR), and marketing that have existed
    for decades. Artificial intelligence is the most overused keyword in
    contemporary solutions. It is typically advertised as a revolutionary approach
    to fulfilling labour demands. Businesses throughout the globe now have access
    to better, cheaper labour as a result of technological improvements.
    Additionally, they can use analytics to identify untapped sources of value for
    their own organisations. Today, artificial intelligence has an undeniable impact
    on business. AI has multiple uses in the business world, including predictive
    analytics for corporate intelligence, deep learning applications for picture
    identification, and recommendation algorithms for the creation of personalised
    suggestions. AI will impact non-technical parts of the economy as well. We will
    study the potential ethical ramifications of implementing AI, ML, and cloud
    technologies in the marketing, finance, healthcare, and human resources
    industries. In addition, we will cover why AI is crucial in business today and
    how these solutions operate.

    See patent

  • ARTIFICIAL INTELLIGENCE AND COMPUTER VISION BASED DETECTION AND PREVENTION OF BRAIN DISORDER DISEASE AND TUMOR AT EARLY STAGE AND TO PROVIDE EFFICIENT TREATMENT USING DATA MINING AND DEEP LEARNING ALGORITHM FOR SMART HEALTH CARE MANAGEMENT

    Issued 202221047801

    Effective medical diagnosis of a wide range of disorders is currently a
    significant unmet need on a global scale. The difficulty of establishing an early
    diagnosis tool and an effective therapy is greatly increased by the complexity of
    the various disease mechanisms and underlying symptoms that are
    experienced by the patient population. Researchers, medical professionals, and
    patients alike now have the ability to overcome some of these problems thanks
    to a subfield of…

    Effective medical diagnosis of a wide range of disorders is currently a
    significant unmet need on a global scale. The difficulty of establishing an early
    diagnosis tool and an effective therapy is greatly increased by the complexity of
    the various disease mechanisms and underlying symptoms that are
    experienced by the patient population. Researchers, medical professionals, and
    patients alike now have the ability to overcome some of these problems thanks
    to a subfield of artificial intelligence called machine learning (ML). This
    overview, which is based on recent relevant research, describes how machine
    learning (ML) is currently being utilised to assist in the early diagnosis of a
    wide variety of diseases. At the outset, a bibliometric examination of the
    publication is performed by utilising information from the Scopus and Web of
    Science (WOS) databases. The bibliometric analysis of 1216 publications was
    carried out in order to identify the most prolific authors, countries, and
    organisations, as well as the papers that were cited the most. Following this, a
    summary of the most current developments and strategies in machine
    learning-based disease diagnosis (MLBDD) is presented. This overview takes
    into account the following aspects: algorithm, illness kinds, data type,
    application, and evaluation metrics. In conclusion, this study focuses on the
    most important outcomes and offers some insight into potential future
    developments and prospects in the MLBDD field.

    See patent

  • DEEP LEARNING BASED TECHNIQUE TO GAUGE BETWEEN STRENGHTS AND WEAKNESSES OF AN EMPLOYEE IN HANDLING EMOTIONAL SITUATIONS

    Issued 202231031340

    Deep Learning based technique to gauge between Strengths and Weaknesses of an
    employee in handling Emotional Situations is the proposed invention. The invention
    aims at studying the strength and weakness of employees of an organization. The
    invention focuses on identifying the capability of the employee in balancing both the
    work and emotional circumstances.

  • IoT, Blockchain Enabled Verifiable Searchable Encryption with Aggregating Authorization using machine learning techniques

    Issued 202241030707

    The proposed method is capable of
    accurately detecting the breach that occurred within the IoT network. At first,
    the method that is now based on blockchain was used to put into action the
    technology that protects users' privacy. Because of its value and significance,
    protecting patient health records (PHR) is the most important part of
    cryptography that takes place over the internet. This is especially true in the
    context of the Internet of Medical Things (IoMT). The…

    The proposed method is capable of
    accurately detecting the breach that occurred within the IoT network. At first,
    the method that is now based on blockchain was used to put into action the
    technology that protects users' privacy. Because of its value and significance,
    protecting patient health records (PHR) is the most important part of
    cryptography that takes place over the internet. This is especially true in the
    context of the Internet of Medical Things (IoMT). The search keywords access
    mechanism is one of the typical methods that is used to retrieve PHR from a
    database; however, it is vulnerable to a variety of security flaws due to the
    nature of the mechanism.

  • FARMERS ATTITUDES, PERCEPTIONS AND EXPECTATIONS OF THE SYSTEM AND SERVICES OF THE REGULATED MARKETS

    Issued 202241030384

    The present invention relates to to integrate India's farming and marketing systems with global
    systems, taking into account not only the current performance of regulated markets, but also the
    changing ways and means of improvising the system according to the attitudes perceived by farmers
    and traders compared to the system with employee participation. This study analyzes this problem.

  • AN AI & ML BASED SYSTEM FOR FABRICATING POWER VLSI DIODE DEVICES AND METHOD THEREOF

    Issued 202141051706

    The present invention discloses an AI & ML based system for fabricating power VLSI diode devices and method thereof. The method and system includes, but not limited to, a VLSI
    compatible substrate; a buried layer provided in the VLSI compatible substrate; a cathode region in the VLSI circuitry, further comprising a high-voltage lightly doped drain in the VLSI
    compatible substrate; a well surrounding high-voltage lightly doped drain and a doping region in the well and surrounding the…

    The present invention discloses an AI & ML based system for fabricating power VLSI diode devices and method thereof. The method and system includes, but not limited to, a VLSI
    compatible substrate; a buried layer provided in the VLSI compatible substrate; a cathode region in the VLSI circuitry, further comprising a high-voltage lightly doped drain in the VLSI
    compatible substrate; a well surrounding high-voltage lightly doped drain and a doping region in the well and surrounding the high-voltage lightly doped drain communicatively coupled with
    an anode region in the VLSI circuitry surrounding the cathode region in the VLSI circuitry; and a guard ring surrounding the anode region and connected to the buried layer

Projects

  • IMAGE CLASSIFICATION USING CIFAR-10

    - Present

    The CIFAR-10 dataset is a collection of images that are commonly used to train machine learning and computer vision algorithms. It is one of the most widely used datasets for machine learning research. The CIFAR-10 dataset contains 60,000 32x32 color images in 10 different classes. The 10 different classes represent aeroplanes, cars, birds, cats, deer, dogs, frogs, horses, ships, and trucks. There are 6,000 images of each class.

    See project

  • DIABETIC RETINOPATHY DETECTION USING DEEP LEARNING

    -

    The purpose of
    the proposed work is to detect diabetic retinopathy in which clinical imaging mode
    is used for analyzing the eye health of the patient. The classification is based on No
    DR, mild Non-Proliferative DR, moderate Non-Proliferative DR, severe NonProliferative DR and proliferative DR using real-world datasets of DR images. A
    comparative analysis is done with various deep learning models like Convolution
    Neural Network, MobileNetv2, ResNet50, Inceptionv2, and…

    The purpose of
    the proposed work is to detect diabetic retinopathy in which clinical imaging mode
    is used for analyzing the eye health of the patient. The classification is based on No
    DR, mild Non-Proliferative DR, moderate Non-Proliferative DR, severe NonProliferative DR and proliferative DR using real-world datasets of DR images. A
    comparative analysis is done with various deep learning models like Convolution
    Neural Network, MobileNetv2, ResNet50, Inceptionv2, and VGG-16. The
    proposed model can also be used to make predictions and find accuracy using less number of images.

    See project

  • WEB DEVELOPMENT BOOTCAMP

    -

    Web Development BootCamp Hi I made this project during the 7 Days Free Bootcamp, conducted by SHAPEAI . The instructor during the session was Mr. Shaurya Sinha (a Data Analyst at Jio). I got to learn a lot during these 7 days and it was an amazing experience learning with SHAPEAI.
    I got to have hands on experience on:

    HTML
    CSS
    during these 7 days, and everything was explained from the very basics so that anyone with zero experience on programming can learn. I enjoyed these 7…

    Web Development BootCamp Hi I made this project during the 7 Days Free Bootcamp, conducted by SHAPEAI . The instructor during the session was Mr. Shaurya Sinha (a Data Analyst at Jio). I got to learn a lot during these 7 days and it was an amazing experience learning with SHAPEAI.
    I got to have hands on experience on:

    HTML
    CSS
    during these 7 days, and everything was explained from the very basics so that anyone with zero experience on programming can learn. I enjoyed these 7 days, you can as well. To register for next free 7 days bootcamp, visit: www.shapeai.tech or follow SHAPEAI on:

    See project

  • MNIST

    -

    MNIST ("Modified National Institute of Standards and Technology") is the de facto “hello world” dataset of computer vision. Since its release in 1999, this classic dataset of handwritten images has served as the basis for benchmarking classification algorithms. As new machine learning techniques emerge, MNIST remains a reliable resource for researchers and learners alike.Each image is 28 pixels in height and 28 pixels in width, for a total of 784 pixels in total. Each pixel has a single…

    MNIST ("Modified National Institute of Standards and Technology") is the de facto “hello world” dataset of computer vision. Since its release in 1999, this classic dataset of handwritten images has served as the basis for benchmarking classification algorithms. As new machine learning techniques emerge, MNIST remains a reliable resource for researchers and learners alike.Each image is 28 pixels in height and 28 pixels in width, for a total of 784 pixels in total. Each pixel has a single pixel-value associated with it, indicating the lightness or darkness of that pixel, with higher numbers meaning darker. This pixel-value is an integer between 0 and 255, inclusive.

    The training data set, (train.csv), has 785 columns. The first column, called "label", is the digit that was drawn by the user. The rest of the columns contain the pixel-values of the associated image.

    Each pixel column in the training set has a name like pixelx, where x is an integer between 0 and 783, inclusive. To locate this pixel on the image, suppose that we have decomposed x as x = i * 28 j, where i and j are integers between 0 and 27, inclusive. Then pixels is located on row i and column j of a 28 x 28 matrix, (indexing by zero).

    See project

  • TITANIC SURVIVORS ANALYSIS

    -

    See project

  • ANALYSIS OF COVID-19 AND PNEUMONIA DETECTION IN CHEST X-RAY IMAGES USING DEEP LEARNING

    -

    COVID-19 (coronavirus) is a global healthcare issue all over the world. Recent stats indicate that more than 29.6 million confirmed cases are diagnosed in international concern as of September 2020. The proposed work aims to detect COVID19 patients and pneumonia patients from X-rays which is one of the medical imaging modes to analyze the health of patients lung inflammation. In the proposed analysis, a comparative study will be done to select the suitable Convolutional Neural Network Model for…

    COVID-19 (coronavirus) is a global healthcare issue all over the world. Recent stats indicate that more than 29.6 million confirmed cases are diagnosed in international concern as of September 2020. The proposed work aims to detect COVID19 patients and pneumonia patients from X-rays which is one of the medical imaging modes to analyze the health of patients lung inflammation. In the proposed analysis, a comparative study will be done to select the suitable Convolutional Neural Network Model for the identified datasets. It is used to detect COVID-19 patients and pneumonia patients on the real-world dataset which could be identified using lung X-rays. Images are preprocessed and trained for various classifications like normal, COVID-19, and pneumonia. After preprocessing, the detection of the disease is done by selecting the appropriate features from the images in each of the datasets. The result indicates the accuracy of detection of COVID vs Normal and COVID vs pneumonia. Among those

    See project

  • CT SCAN IMAGES PREDICTION USING COVID-19

    -

    See project

  • TEXT SUMMARIZATION USING STORIES

    -

    This paper presents the rule-based approach to identify the characters from the story and also determine the Character Extraction among them. This approach can follow basic steps such as Tokenization, POS tagging, sentence parsing followed by the pronoun resolution implementing various TextRank algorithms, and finally extracting character entities among them. The proposed work done can act as a base for story Summarization. Characters are the main essence of any story. Characters and their…

    This paper presents the rule-based approach to identify the characters from the story and also determine the Character Extraction among them. This approach can follow basic steps such as Tokenization, POS tagging, sentence parsing followed by the pronoun resolution implementing various TextRank algorithms, and finally extracting character entities among them. The proposed work done can act as a base for story Summarization. Characters are the main essence of any story. Characters and their descriptions help users to picture the context of the story clearly and more precisely. The extracts character from the stories are used to define weights for the sentences and further rank is based on the importance and similarity of the sentences to each other using extractive methods is retain the most important to summarize the articles.

  • IDENTIFYING HIDDEN CORRELATION OVER FUSION TECHNOLOGY APPLIED IN WIRELESS SENSOR ENVIRONMENT

    -

    Wireless sensor networks (WSNs) have been traditionally tasked with single applications, but now emerged with multiapplication paradigms. As the number of applications in a WSN increase, it also
    increases the WSN complexity and the amount of required transmitted messages. A major requirement
    in these networks is to save energy in order to extend their operational lifetime. Multi-sensor data fusion (MDF) is one of the most widely used. MDF has to identify (hidden) correlations between…

    Wireless sensor networks (WSNs) have been traditionally tasked with single applications, but now emerged with multiapplication paradigms. As the number of applications in a WSN increase, it also
    increases the WSN complexity and the amount of required transmitted messages. A major requirement
    in these networks is to save energy in order to extend their operational lifetime. Multi-sensor data fusion (MDF) is one of the most widely used. MDF has to identify (hidden) correlations between sensors and exploit such knowledge to monitor the behavior of sensors during their working life. And it also
    combines the data into hierarchical form to reduce traffic and energy utilization. Those correlations in a
    multi-application environment could indicate that different applications may share similarities in terms
    of sense that could be used in MDFs to achieve better results in terms of energy consumption and
    MDFs accuracy.

Organizations

  • IEEE

    Member

    - Present

  • Computer Society of India

    Member

    -

View Kanakaprabha’s full profile

  • See who you know in common
  • Get introduced
  • Contact Kanakaprabha directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses

See all courses