Rutgers ECE Student collaborates with Siemens on Biomedical Research Project

Rutgers’ graduate student Sushil Mittal has collaborated with a group of researchers from Siemens Corporate Research and Siemens Healthcare on a biomedical research project. Under the tutelage of Professor Peter Meer, Mr. Mittal spearheaded the project while interning at Siemens Corporate Research in Princeton, NJ. Mr. Mittal is the first author on the first published paper, “Fast Automatic Detection of Calcified Coronary Lesions in 3D Cardiac CT Images*,” to feature research from the project.

The project aimed to address some of the difficulties related to detecting calcified coronary lesions. Despite recent advances in multi-detector computed tomography (MDCT), detection of coronary lesions remains a troubling task. There are multiple reasons why detection is so difficult. For one, the lesions are incredibly small. Secondly, coronary arteries are by nature long and winding, which makes identifying lesions difficult. Finally, the imaging data tends to be of a low resolution, and vulnerable to noise, blooming, and motion artifacts. In response to these issues, Mr. Mittal and the rest of the team have proposed a solution: a novel learning-based, fully automatic algorithm for the detection of calcified lesions in contrast enhanced 3-D CT data.

With guidance from his manager at Siemens and Professor Meer, Mr. Mittal was solely responsible for the implementation and execution of the project experiments. The experiments compared and evaluated the performance of two supervised learning methods—Probabilistic Boosting Tree (PBT) and Random Forest (RF) classifier. Both PBT and RF use rotation invariant features that are extracted along the centerline of the coronaries. On data collected from 165 patients, Mr. Mittal was able to achieve an approximate 90% detection rate for less than one false positive scan. Below are two detection results found on coronary images:

scan-02 (1).jpg

                              Original                                                                     PBT                                                                     RF

Siemens Corporate Research funded the project and Mr. Mittal is currently waiting on the review decision of the project’s second paper. He works with Professor Meer in the Robust Image Understanding Laboratory at Rutgers.

S. Mittal, Yefeng Zheng, B. Georgescu, F. Vega-Higuera, Shaohua Kevin Zhou, P. Meer, D. Comaniciu. Published in Machine Learning in Medical Imaging, Volume 6357 of Lecture Notes in Computer Science, and available online

By Sean Patrick Cooper

Mobility First

The National Science Foundation has awarded a three-year, $7.5 million grant to a Rutgers-led research team to develop a future Internet design optimized for mobile networking and communication.
The team of nine universities and several industrial partners has dubbed its project "MobilityFirst", reflecting the Internet's evolution away from traditional wired connections between desktop PCs and servers toward wireless data services on mobile platforms.

The group will design a "clean-slate" network architecture to accommodate the shift of Internet traffic to smart cellular phones, tablet computers and emerging mobile data services, said Dipankar Raychaudhuri, professor of electrical and computer engineering and director of Rutgers Wireless Information Network Laboratory.

There are more than four billion mobile devices in use worldwide today, and experts predict that by 2015, these wireless devices will significantly outnumber wired devices on the Internet.

"The mobile Internet will do much more than support today's impressive lineup of smart cellular phones. It will simplify peoples interactions with their physical world", Professor Raychaudhuri said. For instance, he said, it will enable location-aware computing, allowing people to find nearby merchants or get driving or public transit directions, even if they don't know their location. It also will support machine-to-machine communications, such as wearable devices that monitor your health and communicate with hospitals or cars that alert other cars to congestion and send split-second commands to each other to avert collisions.

For more information read the complete article at

Sensors aim to monitor smoker activity

The University's Center for Autonomic Computing developed a wireless sensor project that detects human motion and can further medical research.

The sensors, which are small devices that attach to the body, contain accelerometers and gyroscopes that measure movement and can tell what action a person is doing, said Alex Weiner, a School of Engineering junior who is fine-tuning the algorithm of the sensors.

Dario Pompili, assistant professor in the Department of Electrical and Computer Engineering, said the project could help behavioral scientist Theodore Walls from the University of Rhode Island with his research into smoking habits.

Pompili said smokers may not give an accurate self-report on their smoking habits, so doctors can rely on the sensors to give a better report, which can ultimately result in better care for the patient.

"There is a lot of bias in self-reporting. Maybe the smoker smokes more because of the stress of the self-reporting, or he reports a lower amount," he said.

A smoker would need two wireless sensors — one on the wrist and the other on the shoulder — for a computer to understand when and for how long they were smoking, Pompili said.

"The accelerometer captures motion on all three axes, and the gyroscope measures angular velocity," he said.

The sensor project is an extension of former graduate student John Paul Varkey's research in monitoring smoker's actions, Weiner said. The sensor can detect the difference between an arm in the resting position and one raised to the mouth while smoking.

The computer is programmed with a supervised learning algorithm, in which the computer begins to recognize inputs it is fed, Weiner said.

"You tell the machine, ‘This is what smoking looks like,' three or four times [and] next time it should recognize it," he said.

Weiner said the sensor has been successful in recognizing different motions and distinguishing actions like smoking from walking.

"I'd say the success rate is 98 percent for recognizing different actions but only moderately successful in recognizing similar motions," he said.

The sensor recognizes actions with a larger difference in movement easier than actions that are similar, Weiner said.

"Something like brushing one's teeth is a somewhat similar motion to smoking, and the computer would have a hard time distinguishing that," he said.

Although the accelerometer and gyroscopes measure the duration one is smoking for, it does not measure the reasons for smoking, Pompili said.

"We need to understand behavior, we need to [know] when and where you smoke, or if you smoke because you are with someone," he said.

The sensors can measure other aspects besides movement, said Hariharasudhan Viswanathan, a School of Engineering graduate student who contributes to the project.

"We have sensors to measure temperature and humidity, and we also have EKG or ECG sensors that can be used to measure heart rates," he said.

In the future, the project will include a smoke sensor, Viswanathan said.

"Right now we only measure how long the person smokes, but with this we can see if he takes a lot of puffs, and see how much intake there is," he said.

Viswanathan hopes to create an e-doctor application, in which doctors could provide remote health care to patients.

"Let's say you need an EKG. You place three sensor nodes on the body and then the results can go to your doctor. It's cheap and efficient," he said.

Weiner said the project could assist athletes, help doctors in third world countries and monitor the elderly.

Pompili also suggested that sensors could be used in the military for measuring soldiers' vital signs on the battlefield.

"It would be a great application for the army," he said. "Through an ad-hoc wireless network, you could have vital information of the soldiers."

Pompili said the soldiers could wear a smart suit with the sensors on the field, which would feed information back to the base and provide information that was previously unavailable.

"You could read the vitals and predict fainting due to stress and many other problems," he said. "Plus it's a non-invasive way to obtain this information."

This article was published by The Daily Targum

Researchers develop ‘smarter’ smartphone based on personality (ABC News)

Janne Lindqvist's recent work on smartphone interruptions has received a lot of attention in the popular media. The work done with his PhD students Fengpeng Yuan (CS) and Xianyi Gao (ECE) will be presented at CHI’17, the premier tier-1 publication venue for human-computer interaction.


The full paper is available here:

and a 30 second video preview here:

Prof. Lindqvist's work was highlighted on scientific websites such as the NSF, AAAS and ACM, has also received worldwide coverage in widely circulated articles. Below are a few examples:



Fox News:

"Tired of annoying phone alerts? New system could act as a 'secretary' to predict when you want to be left alone", in the Daily Mail

"Pardon the interruption: Here's how your smartphone could be less of a noodge", in Network World

"Smartphone notifications driving you crazy? This might help", in the Economic Times

Congratulations to Janne on the wide attention for his work!

Janne Lindqvist's research featured in "Rutgers Today"

Smartphone Interruptions: Are Yours Relentless and Annoying?  A Rutgers study, featured in Rutgers Today, reveals that personality traits influence and help predict receptiveness to smartphone notifications."Ideally, a smartphone notification management system should be like an excellent human secretary who knows when you want to be interrupted or left alone", said Janne Lindqvist, an assistant professor in the Department of Electrical and Computer Engineering in Rutgers’ School of Engineering. “We know that people struggle with time management all the time, so a smartphone, instead of being a nuisance, could actually help with things.” Read the complete article here

Vishal Patel's work on De-Raining featured in The Outline

Professor Vishal Patel's recent work on de-raining is featured in the publication The Outline. Along with his PhD students He Zhang and Vishwanath Sindagi, Professor Patel has recently developed an algorithm for de-raining rainy images using a deep learning method based on conditional generative adversarial networks. Please read the article entitled "Computers are learning how to see in the rain" at

Elastic Pathing Research featured in IEEE Spectrum, Communications of the ACM, YouTube, Rutgers Today, and MIT Technology Review

Prof. Janne Lindqvist's research shows that how fast you drive might reveal exactly where you are going.   Dr. Lindqvist's work on Elastic Pathing is featured in Rutgers Today

Prof. Lindqvist's Elastic Pathing research has been featured in many publications and news media including YouTube and in MIT Technology Review

Dr. Lindqvist's research is featured in the Communications of the ACM

and the front page of the IEEE Spectrum


Prof. Janne Lindqvist, an assistant professor of electrical and computer engineering, member of WINLAB and director of the Human-Computer Interaction Laboratory, led a team to show how just your driving speed can be used to track where you drive.   This work, "Elastic Pathing: Your Speed is Enough to Track You" is part of a NSF-funded project for which Prof. Janne Lindqvist is the sole Principal Investigator.  

Prof. Lindqvist's paper can be found at and the YouTube video can be accessed by clicking below.


Prof. Janne Lindqvist's team included former PhD student  Dr. Berhard Firner, with ECE PhD students Yulong Yang and Xianyi Gao, recently graduated Master's student Shridatt Sugrim and undergraduate student Victor Kaiser-Pendergast.

The motivation for the project was that today people increasingly have the opportunity to opt-in to "usage-based" automotive insurance programs for reducing insurance premiums. In these programs, participants install devices in their vehicles that monitor their driving behavior, which raises some privacy concerns. Some devices collect fine-grained speed data to monitor driving habits.


Prof. Lindqvist's Gesture Security work featured on RU-TV and many other news and media outlets

Prof. Janne Lindqvist's interdisciplinary security work which was presented at MobiSys'14, the tier-1 conference on mobile systems, is getting nice publicity around the world.

In that work, Prof. Janne Lindqvist and his students, Michael Sherman (former ECE undergrad, and currently WINLAB staff), Gradeigh Clark (ECE PhD student), Yulong Yang (ECE PhD student) and Shridatt Sugrim (ECE MS and WINLAB staff), collaborated with Prof. Antti Oulasvirta of Max-Planck Institute for Informatics and Teemu Roos of University of Helsinki, on a novel form of authentication for mobile devices. In particular, they studied user-generated free-form gestures and developed a novel information-theoretic for analyzing the security and memorability of the gestures. The group also built an actual authentication system designed for the gestures.

On October 6th, Dr. Lindqvist's research was featured on "Wake Up Rutgers", a daily Rutgers TV program. The show is available online at the link below. Dr. Lindqvist's segment starts at 23 minutes and ends at 26 minutes.

The research paper is available at:

A press release including a video made can be found at

The work has been featured on   CBS Radio News   and appeared in the following outlets:


Scientific American

Front page of NSF's 360 degrees web site at the moment:

Yahoo! News

International Business Times, which reaches 5 million people in the UK and 50 million around the world

Science Daily

Front page of Scientific Computing

2nd largest daily newspaper in the UK

R&D Magazine

A lot of coverage in India including major publishers including major publishers

Established in November 1981, it is the oldest and most widely circulatedEnglish-language broadsheet in Oman:

And lots more..


Subscribe to Rutgers University, Electrical & Computer Engineering RSS