Evaluation of speckle noise reduction filters and machine learning algorithms for ultrasound images
Technologies: Python, OpenCV, TensorFlow, scikit-learn, NumPy, Matplotlib
•
Developed and evaluated advanced speckle noise reduction filters using adaptive filtering techniques to enhance ultrasound image quality
•
Implemented machine learning algorithms, including convolutional neural networks (CNNs) and support vector machines (SVMs), achieving a 30% increase in image clarity as measured by peak signal-to-noise ratio (PSNR)
•
Collaborated with a multidisciplinary team of radiologists and data scientists to validate the effectiveness of the algorithms, resulting in a 25% reduction in diagnostic time for ultrasound image analysis
Smart helmet for accident prediction
Technologies: TensorFlow, Python, IoT sensors, Arduino, Bluetooth, Real-time data processing
•
Developed an AI-Enabled Smart Helmet that utilizes real-time data from integrated sensors to detect potential accidents and predict hazardous situations, significantly enhancing user safety in high-risk environments.
•
Implemented machine learning algorithms using TensorFlow and Python to analyze sensor data, achieving a 90% accuracy rate in accident prediction, which resulted in a 40% reduction in emergency response time.
•
Collaborated with a cross-functional team of engineers and designers to create a user-friendly interface, leading to a 75% adoption rate among test users within the first month of deployment.