You are on page 1of 1

HEALTHCARE ROBOTICS

Sensory Computing and Object Processing Entities


John Bachkosky, Alexandra Boukhvalova, Kevin Chou, William Gunnarsson,
James Ledwell, Brendan McTaggart, Xiaoqing Qian, Nicholas Rodgers, John Shi, Jason Yon
Mentor: Dr. Anil Deane Librarian: Elizabeth Soergel

Introduction Healthcare Companion Robot - ALVORES Mobility Platform


The healthcare industry is being revolutionized by computer-
based technologies, the newest and likely most influential of Healthcare Applications:
which is artificial intelligence (AI), which can assist in patient Medication Retrieval
consultation, diagnostics, treatment, and other functions1. AI
systems can even improve healthcare operations, as they can
efficiently perform remedial tasks and accurately recommend Disease Recognition
diagnoses or treatments using information stored in large
databases2,3. Team SCOPE will advance the state of AI in Patient Vital Monitoring
healthcare by using object recognition and natural language
processing techniques to create a medical personal assistant.
Doctors Assistant
Raspberry Pi (blue box) and Arduino (red box) communicate between each
Personal Healthcare
Object Recognition other and the Android tablet to control and move ALVORES. ALVORES
uses proximity sensors (orange box) and motors attached to each wheel
(yellow box) in order to efficiently move around.
Image acquisition Patient Information Database
The picture is taken without requiring user input. The platform will steer autonomously towards a desired object.
Setting the image Front view of ALVORES, Version 1. Shows Android Platform operating according to external input from Android
Tablet running Object Recognition Application. tablet.
The picture is cut into n pieces, called segments.
Uploading the segments The platform has four active states: forwards, backwards, turn
The segments are sent to IBMs Bluemix Services via the left, and turn right.
Alchemy API. Component Integration Proximity sensors interrupt movement when detecting collisions
Bluemix returns phrases describing each given segments. Platform will use a rack and pinion steering system

Timeline to Thesis Conference

Object recognition software run on three images with the Bluemix output
presented to the right of the input image. Time measurements detail how
quickly IBM Bluemix can identify each image.

References
Natural Language Processing 1. Luxton, D. D. (2015). An Introduction to Artificial Intelligence in Behavioral and Mental
Health Care. Artificial Intelligence in Behavioral and Mental Health Care, 1.
Goal: Develop upon already existing software structures and 2. Herrero-Zazo, M., Segura-Bedmar, I., Hastings, J., & Martnez, P. (2015). Application of
technologies: Natural Language Toolkit (NLTK) and IBMs Bluemix Text Domain Ontologies to Natural Language Processing: A Case Study for Drug-Drug
Interactions. International Journal of Information Retrieval Research (IJIRR), 5(3), 19-
Processing. 38.
3. Tan, H., Xu, Y., Mao, Y., Tong, X., Griffin, W. B., Kannan, B., & DeRose, L. A. (2015,
Already Completed / In Future Work May). An integrated vision-based robotic manipulation system for sorting surgical tools.
In Technologies for Practical Robot Applications (TePRA), 2015 IEEE International
Development Expand database capabilities Conference on (pp. 1-6). IEEE.
Command, Statement, & Incorporate visually collected
Query Processing Programs information from robot into database
Simple database for Robot capability to answer questions
Acknowledgements
Text processing through NLTK software on
statement storage and complete tasks based on prior Special Thanks to: Dr. Anil Deane, Ms. Elizabeth Soergel, Dr. Frank
commands and statements. Verification on
human interaction type of query and part of speech tags are Coale, Dr. Kristan Skendall and the Gemstone Program Staff.
displayed.

You might also like