ZANN LIM JIA MIN
ML-based Breast Ultrasound Classification for Early Cancer Detection
Breast cancer diagnosis accuracy is critical for early intervention and improved patient outcomes. Our project presents an ML model for classifying breast ultrasounds into normal, benign, and malignant categories, streamlining tumor detection and enhancing diagnostic efficiency. Targeting healthcare professionals, medical imaging researchers, and breast cancer patients, our model provides a reliable second opinion. The model also reduces human error, and aids in identifying subtle patterns in ultrasound images. Integration with information systems facilitates image analysis for medical professionals and researchers. By enabling accurate and timely detection of early-stage breast cancer, our project contributes to more effective treatments and better patient outcomes.
CoPr: In-Context Learning and Prefixes Tuning for Grammar Error Correction (GEC)
Communication is a huge part of our way of life, grammatical errors can sometimes lead to misunderstandings while communicating especially in the absence of body languages. Credibility of the text may also seem less credible and impactful when the text is filled with grammatical errors.