Watjana Lilaonitkul

Title: Watjana Lilaonitkul: Medical Imaging and Neural-Network Inventions.

Introduction

Watjana Lilaonitkul is an inventor based in London, Great Britain. His work includes inventions applying neural networks to medical image analysis and object detection. He is associated with UCL Business Plc and has 2 patents listed in the information provided.

Latest Patents

One of Lilaonitkul’s listed inventions, “Medical image analysis using neural networks,” describes systems and methods for identifying layer structures in medical images. A processing device uses a neural network to assign each pixel a boundary-surface score, indicating the likelihood that it lies between segmented tissue layers. The described approach can use two connected neural-network sub-networks: one identifies tissue layers, while the other determines the boundaries between them. The first sub-network may include multi-scale pooling to process image features at different sampling rates.

His other listed invention, “Computer-implemented method of enhancing object detection in a digital image of known underlying structure, and corresponding module, data processing apparatus and computer program,” focuses on improving object detection. The method uses positive images containing detected objects and negative images showing the same underlying structure without detected objects. A neural network extracts feature vectors from image pairs, contrasts those vectors to create an attention vector, and processes that vector alongside features from the positive image to produce an output.

Career Highlights

The inventions described here demonstrate applications of neural networks to two image-processing challenges: identifying tissue boundaries in medical imagery and enhancing object detection in images with known underlying structures. Lilaonitkul’s listed affiliation is UCL Business Plc.

Collaborations

The provided information names Adam Dubis and James Willoughby as Lilaonitkul’s coworkers. It does not specify their roles in the listed inventions.

Conclusion

Watjana Lilaonitkul’s listed patent work applies neural-network techniques to medical image segmentation and object detection, reflecting a focus on computational methods for analyzing complex images.