•Spearheaded a pioneering project on teeth segmentation and periapical pathology detection using orthopantomograms, achieving a significant improvement in detection accuracy and analysis speed.•Compiled and curated a comprehensive dataset of orthopantomograms, including image preprocessing and augmentation techniques to enhance model training effectiveness and robustness.•Innovated with Attention U-Net models to refine segmentation quality, enabling precise identification of dental issues and enhancing diagnostic reliability.•Advanced the application of DeepLabv3+ in dental radiography, optimizing the model for superior pathology detection and segmentation in complex images.•Collaborated cross-functionally with radiologists and dental professionals to validate models and integrate clinical insights, ensuring high relevancy and applicability of solutions.•Conducted extensive testing and validation, demonstrating the models' effectiveness in real-world scenarios.

•Developed a predictive maintenance model for active machines using python, python flask, python libraries like (Pandas, Numpy, seaborn, matplotlib, SkLearn, Time series), ESP32 and mongo db. •The research focused on studying various types of machines and their associated failure patterns for predictive maintenance.•Done some IOT work, like utilized ESP32 module for data collection, leveraging its capabilities for sensing and transmitting data from connected devise.•The collected data was stored in MongoDB database, allowing for efficient storage and retrieval.•Additionally, developed API and generated API key using Python Flask, enabling seamless communication and interaction with IoT system.•Analyzed data from sensors and other sources to detect faults and threats before they cause failure.•Done data processing like cleaning the data.•Conducted EDA using visualization techniques such as histogram, scatter plot, pair plots, and heatmaps to provide valuable insights into data distribution, relationships, and correlations.•Utilized SkLearn to create a machine learning pipeline that trains on historical data and predict failures.•Used machine algorithm (Logistic regression, random forest, gradient boosting, SVM, K-NN, etc.) •Designed an intuitive user interface that displays real time analytics and alerts to help maintenance teams take timely and informed decisions. •Optimized the model by fine-tuning hyperparameters and evaluating its accuracy and performance metrics. •Successfully deployed the model to production, resulting in reduced downtime, improved safety and significant cost saving for the company.

Car Price Prediction.•Conducted research on the car market, collecting data on car prices, types, owners, companies, and years to analyze trends and factors affecting car values.•Cleaned and preprocessed the data to remove missing values, outliers, and inconsistencies and transformed variables as needed.•Applied different types of machine learning models such as random forest, linear regression, gradient boosting to predict car prices based on features such as make, model, year and condition.•Conducted feature selection and engineering to identify the most important variables and improves model performance.•Evaluated model accuracy using metrics such as mean squared error, R squared and accuracy and compared performance across different models.House Price Prediction.•In the context of a housing price prediction project using a California house dataset from Kaggle, I conducted extensive data processing and exploration. •Leveraging visualization techniques such as histograms, pair plots, scatter plots, and box plots, I gained valuable insights into the dataset's characteristics and relationships. •To predict the mean value of houses in different California locations, I employed machine learning algorithms including Linear Regression, Gradient Boosting, and Random Forest.•I assessed the performance of these algorithms by calculating their accuracy and mean square error. Through rigorous evaluation, I determined the effectiveness of each algorithm in predicting housing prices. •This comprehensive analysis allowed for data-driven decision-making and provided valuable information for further model selection and refinement.•Overall, this project showcased my expertise in data preprocessing, visualization, and the application of various machine learning algorithms for accurate house price predictions.