Neural Discourse Parsing and Computer Aided Diagnosis for Lung Cancer

We are pleased to have two speakers for two interesting topics. Dr. Attapol Rutherford, a Lecturer at the Department of Linguistics, Faculty of Arts, Chulalongkorn University will give us a presentation on the topic of "Neural Discourse Parsing" and Asst.Prof.Dr. Pinyo Taeprasartsit, a lecturer at the Department of Computing, Faculty of Science, Silpakorn University will present "Computer Aided Diagnosis for Lung Cancer: Where Traditional and Modern Techniques Join Force".

Agenda:

6:00 PM - 6:30 PM | Registration

6:30PM - 7:30 PM | Dr. Attapol Rutherford,
Neural Discourse Parsing

7:30 PM - 8:30 PM | Asst.Prof.Dr. Pinyo Taeprasartsit
Computer Aided Diagnosis for Lung Cancer: Where Traditional and Modern Techniques Join Force

8:30 PM - 9:00 PM | Networking

--------------------------------------------------------
Topic 1 : Neural Discourse Parsing
Speaker: Dr. Attapol Rutherford

Abstract of the talk:
Inferring implicit discourse relations in natural language text is the most difficult subtask in discourse parsing. Many neural network models have been proposed to tackle this problem. However, the comparison for this task is not unified, so we could hardly draw clear conclusions about the effectiveness of various architectures. Here, we propose neural network models that are based on feedforward and long-short term memory architecture and systematically study the effects of varying structures.

Speaker bio: https://attapol.github.io/

--------------------------------------------------------
Topic 2 : Computer Aided Diagnosis for Lung Cancer: Where Traditional and Modern Techniques Join Force
Speaker: Asst.Prof.Dr. Pinyo Taeprasartsit

Abstract of the talk:
Lung cancer is the deadliest form of cancer. In standard practice, to predict survival and suitable treatment, a physician performs biopsy through a bronchoscopic procedure to collect suspect tissues in the lungs and mediastinum. Unfortunately, 3D structures of airway trees are highly complex and blood vessels are obstacles that need to be avoided. Without computer-aided tool, the physician, in fact, executes biopsy blindly and usually cannot sample the suspect tissues (< 25% success rate). To improve both efficiency and effectiveness of the procedure, researchers have proposed a set of algorithms and tools as a system for bronchoscopic planning and guidance to provide vivid information to physicians. The success of the system is derived from combined knowledge of mathematics, electrical and biomedical engineering, algorithms, and machine intelligence.
In this talk, we discuss about the techniques laying down the foundation of this complex, indispensable diagnostic tool for battling against lung cancer. This includes patented global tree optimization algorithm, path optimization, augmented reality 3D visualization, and deep learning. This presentation will also demonstrate the beauty of traditional and modern machine intelligence. While we have intensively employed deep learning, there are many applications that well-formulated traditional methods are better options. Researchers are encouraged to learn both traditional and modern techniques and wisely utilize them to answer the needs.