Neuroconductor was developed in the Department of Biostatistics at Johns Hopkins University by John Muschelli (lead R project expert), Adrian Gherman (lead software engineer), Ciprian Crainiceanu and Brian Caffo (Principal Investigators). Neuroconductor (https://neuroconductor.org/) is an open-source platform for rapid testing and dissemination of reproducible computational imaging software. The goals of the project are to:

provide a centralized repository of R software dedicated to image analysis;
disseminate quickly software updates;
educate a large, diverse community of scientists using detailed tutorials and short courses;
ensure quality via automatic and manual quality controls; and
promote reproducibility of image data analysis.

The SMART research group has deployed multiple free online courses on COURSERA as part of a new initiative to deploy an entire specialization on Neuroimaging. The ideas is to create a complete set of tools and tutorials for introducing important ideas from Neuroimaging and for providing a stress-free on-demand system for novice and advanced Neuroimaging users ... read more

oasis_logo.png

Introducing the oasis R package for MS lesion segmentation

Introducing the oasis R package — a R package implementation of OASIS is Automated Statistical Inference for Segmentation (OASIS), currently hosted on github. The OASIS algorithm takes structural brain MRI studies with a FLAIR, T1, T2, and PD volume from patients with multiple sclerosis (MS) and produces OASIS probability maps of MS lesion presence, which can be thresholded into a binary lesion segmentation. With the oasis R package, you can make OASIS probability maps of lesion presence using the model described in the OASIS paper. You can also retrain an OASIS model using your own data and gold standard lesion segmentations and make predictions from this model. For more information on the OASIS package check out this blog plot post. The package was developed by smarties Elizabeth Sweeney, John Muschelli, and Taki Shinohara.

Ciprian_dean_lect_img2.png

Ciprian's Dean lecture entitled Not everybody, but some people walk like you: Wearable computing in public health.

Ciprian discusses the use of accelerometers, video cameras, GPS devices and other wearables in public health. Special emphasis is given to movement type recognition and association between the circadian rhythm of activity and aging and BMI. You can watch the lecture here.

smartstatstools.gif

First automatic segmentation methods for multiple sclerosis lesions

The smart-stats-tools website houses software for the analysis of structural magnetic resonance imaging (MRI) software developed by members of the SMART group. Currently the website is home to the SuBLIME software. SuBLIME is an automated method for segmenting incident multiple sclerosis (MS) lesion voxels between baseline ... read more

Cartoon picture of Brian

Wearable Computing and Accelerometry for the Uninitiated

The SMART group has an active working subgroup in wearable computing and accelerometry. Here's a cartoon version of Brian explaining some of the relevant work done by the group. Wearable computing has the potential to be a revolution in public health by providing much more accurate measurement of behavior among other things. In addition to the technological problem, wearable computing generates complex data that needs careful thought to be analyzed. The SMART group working in wearable computing is at the leading edge of the statistical analysis of this kind of data. ... read more

neuro-quants.png

WANTED: Neuro-quants

... considering the broad demand for statisticians in this area, we recall the famous quip from Jeff Hammerbacher, a former Facebook engineer, who said, "The best minds of my generation are thinking about how to make people click ads. That sucks." We contend that working in biostatistics using complex neuroimaging data to solve the biggest puzzle of them all--understanding the human brain--is a worthy alternative... read more

npr_figure.png

How Much Does It Hurt? Let's Scan Your Brain

For the first time, scientists have been able to predict how much pain people are feeling by looking at images of their brains The findings, published in the New England Journal of Medicine, may lead to the development of reliable methods doctors can use to objectively quantify a patient’s pain. Now, pain intensity can only be measured based on a patient’s own description, which often includes rating the pain on a scale of one to ten. ... read more

Mathematical Biostatistics Boot Camp

SMART group leading the way in massive open online courses

SMART group cofounder Brian Caffo is leading the Coursera course Mathematical Biostatistics Boot Camp. The course has fifteen thousand students ... read more

SMART members in Heritage Health Competition Team

Several members of the SMART group are competing with team Hopkins Biostat at the Kaggle Heritage Health competition. The goal of the competition is to predict days in hospital from a previous year's insurance claim. The team uses over a million insurance claims to build predictive models. Currently our team's best position is ... read more

Statistical methods

Our philosopy is to start with exciting new scientific problems, identify, and carefully tune statistical analytic approaches designed to solve them. Simple, easy to explain solutions are often the best, while other times subtler approaches are necessary. Not every application leads to a new method and many years of problem solving sometimes result in generalizable expertise. We call that the statistical method, ... read more

BrianCaffo_PECASE_h200.jpg

Brian S. Caffo, co-founder of the SMART research group, receives the Presidential Early Career Award for Scientists and Engineers (PECASE) Award

Brian S. Caffo, co-founder of the SMART research group, receives the Presidential Early Career Award for Scientists and Engineers (PECASE) Award, the first such award for a Statistician. White House press release. School of Public Health press release.

Biosignals (EEG, ECG, ECoG, movement)

Biosignals is a word that we invented to cover a wide range of applications in Statistics. It was originally intended to describe data that are collected from humans and animals using non-invasive sensors, such as electric leads, microphones, video recordings, etc. Once we started analyzing ECoG data, which is obtained directly from hundreds of electrodes sitting on the surface of the brain after the skull is surgically removed, we were compelled to drop the "non-invasive" label. Thus, we currently define a Biosignal to be ... read more

Wearable computing (accelerometers, heart monitors, actigraphs)

Wearable computing refers to a new class of measurement devices that can be attached to the human body to measure human activity, biosignals, or environmental factors in real time. Examples of wearable devices are accelerometers, which can measure acceleartion at the hip or wrist, heart monitors, which measure heart rate or systolic blood pressure, and actigraphs, which measure the number of activities per unit of time. These devices can often transmit signals in real time to a central monitoring ... read more

Brain imaging - Clinical

As part of clinical practice and longterm research protocols, many biomedical researchers acquire large databases of diagnostic images. These images are used as part of prospective observational studies to learn about the etiology of disease as well as for comparisons between disease subtypes and treatments. Large clinical imaging databases provide an unique resource for informing research and treatment. However, the scale of data and the heterogenous nature of their acquisition makes analysis complex and inference difﬁcult ... read more

Brain imaging Prediction

An important application of imaging is to try to predict health outcomes using various brain imaging modalities. The SMART group has many parallel efforts in using brain imaging for prediction. This includes efforts to produce better, more biologically motivated predictors as well as efforts in parsimonious prediction based models ... read more

image_mini.png

SMART research group led by Brian and Ciprian wins the ADHD-200 competition.

The Child Mind Institute (CMI), along with Oregon Health and Science University, today announce the winner of the "ADHD-200 Global Competition." The competition, to develop a predictive tool for ADHD diagnosis based on functional and structural magnetic resonance imaging (MRI) of the brain, is part of a movement towards open science and collaboration fostered by the International Neuroimaging Data-sharing Initiative (INDI). The competition sought to inspire scientists to develop novel approaches to real-world problems. This exciting experiment concluded ... read more

scientific1_small.png

Variability, replication, and longitudinal studies

One of our main areas of research is Statistical methods for the analysis of populations of brain images. Expertise includes fMRI, structural MRI (T1, T2, FLAIR, PD, etc), SPECT, CT. Areas of applications include Attention Deficit Hyperactivity Disorder (ADHD), Multiple Sclerosis (MS), Alzheimer disease (AD), brain cancer, Traumatic Brain Injury (TBI), and Stroke. Our brain imaging collaborators ... read more

Previous Pause Next

Home

This is the webpage for the Statistical Methods and Applications for Research in Technology (SMART) working group of the Department of Biostatistics in the Bloomberg School of Public Health at Johns Hopkins University. The group was started by Ciprian Crainiceanu and Brian Caffo though it now boasts a core group from around the world

About Us

We work on Statistically principled methods for new technologies with special emphasis on brain imaging (e.g. fMRI, high resolution MRI, CT), wearable computing (e.g. hip accelerometers, heart monitors), and Biosignals (e.g. EEG, EKG, ECoG). The underlying principle is to develop methods that are automated, fast, scalable, and robust (AFSR.) Our analytic approaches are sometimes focused only on one subject, but typically we are investigating large populations observed at one or multiple time points.

Contact Us

If you have any questions about the group, research, or website you can contact one of the key personnel. If you want to be added to or deleted from the SMART group e-mail list contact Brian Caffo. If you are a potential collaborator it helps to be highly recommended by one of our scientific collaborators. Well defined ideas, a healthy understanding of the need for Biostatistical thinking and methods, experience in peer-to-peer scientific collaborations are also a big plus.

Search form