Computer vision algorithms used to diagnose depression

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://medvr.ict.usc.edu/projects/dcaps/

SimSensei software, developed by Stefan Scherer and colleagues at the University of Southern California, combines computer vision algorithms and the psychological model of depression. An on-screen psychologist asks you a series of questions and watches how you physically respond. Using Kinect, the computer vision algorithms build up a very detailed model of your face and body, including your “smile level,” horizontal gaze and vertical gaze, how wide open your eyes are, and whether you are leaning toward or away from the camera. From these markers, SimSensei can determine whether you’re exhibiting signs that indicate depression — gaze aversion, smiling less, and fidgeting.

SOINN artificial brain learns from the internet, applies information

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://haselab.info/soinn-e.html

A group at the Tokyo Institute of Technology, led by Dr. Osamu Hasegawa, has advanced SOINN, their machine learning algorithm, which can now use the internet to learn how to perform new tasks. The system, which is under development as an artificial brain for autonomous mental development robots, is currently being used to learn about objects in photos using image searches on the internet. It can also take aspects of other known objects and combine them to make guesses about objects it doesn’t yet recognize.

Neural codes of diseases studied to discover potential “electroceutical” treatments

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://www.reuters.com/article/2013/04/10/us-glaxosmithkline-electroceuticals-idUSBRE9390VM20130410

The Feinstein Institute for Medical Research, the University of Pennsylvania, MIT and GlaxoSmithKline are collaborating on research which aims to discover medicines that use electrical impulses to regulate the body’s organs and functions.

Nearly all organs and functions in the body are regulated through circuits of neurons that communicate through electrical impulses. There already exist devices that use electrical impulses to treat disease (i.e., pacemakers, defibrillators, deep-brain stimulation), but these devices do not target specific cells in the body. Researchers now believe it is possible to create devices that control action potentials in individual neurons, a critical step in developing technologies to use neural circuits to control specific cells. It may be possible to intervene in a broad spectrum of diseases, like inflammatory and autoimmune diseases, because these conditions can be controlled by neurons.

Research links autism to environmental factors

FacebooktwitterlinkedinFacebooktwitterlinkedin

https://imfar.confex.com/imfar/2013/webprogram/Paper14885.html

New studies lend strength to the notion that environmental influences before birth play a role in the risk for the autism.

At the recent International Society for Autism Research annual conference, Marc Weisskopf of the Harvard School of Public Health presented results from a large national study, known as the Nurses’ Health Study II.  The research suggested that a mother’s exposure to high levels of certain types of air pollutants, such as metals and diesel particles, increased the risk of autism by an average of 30% to 50%, compared with women who were exposed to the lowest levels.

Machine learning algorithms analyze mobile phone data for Parkinson’s research

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://mobihealthnews.com/22076/michael-j-fox-foundation-takes-first-step-toward-crowdsourced-research/

The Michael J. Fox Foundation is exploring how data sourced from mobile phones and analyzed with machine learning algorithms can improve Parkinson’s research.  The research was crowdsourced via a public competition.

The initial study included 16 individuals — nine patients, seven control.  For 8 weeks, 4-5 hours per day, each carried a smartphone with seven sensors collecting data. Inputs of a built-in accelerometer, data about the user’s tone of voice, how much the phone was turned on and used, data from the built-in compass and GPS, and an ambient light sensor were analyzed. A machine learning algorithm was developed to use the data to identify the Parkinson’s patients from the control group and identify what stage of the disease users were in.

DNA “nanotrain” for targeted cancer drug delivery

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://news.ufl.edu/2013/04/29/nanotrain/

The nanotrain cost-effectively delivers high doses of drugs to precisely targeted cancers and other medical maladies without leaving behind toxic nano-clutter.

“The beauty of the nanotrain is that by using different disease biomarkers you can hitch different types of DNA probes as the train’s ‘locomotive’ to recognize and target different types of cancers,” said Weihong Tan of the University of Florida. “We’ve precisely targeted leukemia, lung and liver cancer cells, and because the DNA probes are so precise in targeting only specific types of cancer cells we’ve seen dramatic reduction in drug toxicity in comparison to standard chemotherapies, which don’t discriminate well between cancerous and healthy cells.”

Ingestible sensors alert doctors and caregivers when a pill is taken

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://www.forbes.com/sites/nextavenue/2013/04/16/the-newest-high-tech-pill-will-text-when-swallowed/

Proteus Digital Health is creating a new category of products, services and data systems that have the potential to significantly improve the effectiveness of existing pharmaceutical treatments.  Called Digital Medicines, these new pharmaceuticals will contain a tiny sensor that can communicate, via a digital health feedback system, vital information about an individual’s medication-taking behavior and how their body is responding.

Brain scans link math learning abilities to brain structure

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://scopeblog.stanford.edu/2013/04/29/peering-into-the-brain-to-predict-kids-response-to-math-tutoring/

Researchers from the Stanford University School of Medicine used brain scans to look for a link between math-learning abilities and brain structure or function, and compared neural and cognitive predictors of childrens’ responses to tutoring.

The analysis of the children’s structural brain scans showed that larger gray matter volume in three brain structures predicted greater ability to benefit from math tutoring. The predictions were generated with a machine learning algorithm.

The researchers’ next steps will include comparing brain structure and wiring in children with and without math learning disabilities, analyzing how the wiring of the brain changes in response to tutoring, and examining whether lower-performing children’s brains can be exercised to help them learn math.

Babies’ consciousness, development studied

FacebooktwitterlinkedinFacebooktwitterlinkedin

http://news.sciencemag.org/sciencenow/2013/04/when-does-your-baby-become-consc.html

Finding the point at which babies’ reactions change from being purely reflexive to reflecting more intention is leading researches to focus on the first glimmers of conscious thought in infants as young as 5 months old.

Ideally, the infant studies would enable scientists to trace a trajectory of how consciousness generates. “You can start to use this method very early to basically try to check whether there is normal or abnormal development,” says Sid Kouider, a researcher at the École Normale Supérieure in Paris.  “We know that autistic children can have trouble being aware of faces, and you could imagine this kind of method to diagnose early on whether someone is reacting in a normal way to objects or faces.”