Research Project Library
With AIClubProjects, you'll embark on an intellectual voyage that harnesses the power of cutting-edge technologies, including deep learning, Large Language Models, and Generative AIs. Unlike traditional research, our preset projects provide you with a roadmap, ensuring that you navigate through the complex world of AI with purpose and precision.
At AIClub, your research journey is made smoother with experts who have several years of experience in the field.
Drug Discovery Project
Huntingtin (HTT) Protein
This research proposes AI-based models that can potentially detect drug efficacy to effectively inhibit the following diseases. Some of the diseases or conditions associated with Huntingtin overexpression include Huntington's disease, which is an autosomal dominant neurodegenerative disorder characterized by motor, cognitive, and psychiatric symptoms.
Serine/threonine-protein kinase PLK1
This research proposes AI-based models that can potentially detect drug efficacy to effectively inhibit the following diseases. Some of the diseases or conditions associated with Serine/threonine-protein kinase PLK1 overexpression include various types of cancers such as breast, lung, colorectal etc.
Vitamin D Receptor
This research proposes AI-based models that can potentially detect drug efficacy to effectively inhibit the following diseases. Some of the diseases or conditions associated with VDR overexpression include Hypercalcemia, Sarcoidosis, Williams Syndrome, Autoimmune Diseases, Fibrotic Diseases, Osteoporosis or Other Bone Disorders.
Computer Vision Projects
NLP Projects
Bioinformatics Projects
Audio Classification Projects
Data Science Projects
Time Series Projects
Machine Learning in Eye-Tracking Section of the Child Mind Network using EEG Data
This is release 8 of HBN-EEG, the EEG and Eye-Tracking Section of the Child Mind Network Healthy Brain Network (HBN) Project, curated into the Brain Imaging Data Structure (BIDS) format. This dataset is part of a larger initiative to advance the understanding of child and adolescent mental health through collecting and analyzing neuroimaging, behavioral, and genetic data. We can create a classification model to determine which task the child is performing from the tasks given in the dataset.
Object Detection
Preventing Gun Violence using object detection
This project focusses on designing a system to identify firearms in surveillance footage in real-time using advanced artificial intelligence models. It provides a crucial tool for preventing gun violence. By integrating with existing security infrastructure, it offers an efficient and effective solution for enhancing safety across various environments, from schools to public spaces, potentially saving lives and maintaining peace
Star/Constellation detection
Utilizing the "Stars and Constellation Dataset" with 16 constellation classes, the state of the art YOLO
models are trained, fine-tuned, and evaluated for performance across varying hyper-parameters. This research leverages the latest YOLOv8 and YOLOv9
architectures, renowned for their real-time object detection capabilities, to create an automated system for celestial object classification
Object Segmentation
Breast Lesions USG segmentation
This research highlight key insights into their effectiveness in BC detection and provide recommendations based on their application to ultrasound imaging. The findings of this study contribute to the ongoing efforts to improve AI-based diagnostic tools for breast cancer. It uses state of the art segmentation techniques to segment the breast tumors which can assist doctors to find the tumor easily.
Car damage detection and segmentation
This study explores the potential of utilizing pre-trained YOLOv8 and YOLOv9 object detection and image segmentation models from Ultralytics to automate and enhance the auto insurance claim process. Training these models on a dataset of 4,874 vehicle damage images aims to streamline the claims process, reducing the need for manual inspections and speeding up report generation.
Flood water segmentation
Flood-prone communities e.g. coastal areas experience frequent flooding due to storm surge, heavy rain and sea level rise. This research uses a dataset consists of 441 annotated roadway flood images that can be used as training samples to train computer vision based flood segmentation algorithms.
Tabular Data
Teaching methods for autistic students
ASD (Autism Spectrum Disorder) affects children globally, with its prevalence on the rise. Early detection is
crucial for educating affected children. In this study, we explore a machine learning-based approach to identify personalized
teaching methods for children with ASD.
RAG Projects (Retrieval Augmented Generation)
Circuit Analyzer using vision API
This project aims to develop a web application for circuit analysis. The application uses the extensively collected information about circuits, as input to a model, and leverages the RAG (Retrieval-Augmented Generation) framework to generate responses to user queries. A vector space is employed to store the data, enabling the system to retrieve relevant information about the circuit provided as input by the user.
AI-Powered Support System for Special Needs Children Using Adaptive Playbooks
The objective is to develop an AI system that assists special needs children by understanding language, speaking, and answering questions using a predefined playbook. The AI will be trained to provide support based on specific scenarios and documents, promoting inclusive and accessible education.
Medical Projects
Early Predictions of Epilepsy Seizures
Epilepsy is interpreted as a neurological disorder characterized by the occurrences of seizures, due to abnormal excessive or synchronous neuronal activity in the brain. Very importantly, the main features of epilepsy, epileptic seizures, are associated with a number of negative consequences at both short- and long-term, including the risk of falls and injuries, eventually death, psychiatric disturbances, cognitive deficits and difficulties in achieving academic, social and employment goals. Treatment options for epilepsy are pharmacological and surgical. The database consists of EEG recordings of 14 patients. Time series based analysis for seizure predictions dataset. Time series feature extraction based on tsfresh for 15 seconds 5minutes early window.
