Feature of the Day


2024:

  • (20240113) TUSZ (v2.0.2): We have removed two duplicate sessions and fixed one bug in the annotations. Performance on seizure detection should not change significantly.

2023:

  • (20231104) We are looking for new PhD students for the Fall 2024 semester. We have several fully funded positions available. Applicants should hold an MS degree in electrical or computer engineering from an accredited university. Applicants should be interested in doing fundamental algorithm research in machine learning with applications in automatic interpretation of EEGs, digital pathology images, and cardiology signals. To learn more about our active projects, go here. In addition to these technical directions, applicants should be interested in software engineering - we take software development very seriously. We also are involved in production of data - something we consider very important also.

    Applicants MUST be interested in working in a team, contributing to our software and supervising teams of undergraduate students. You should have good communication skills and take an active interest in publishing your work.

    Joining our team can be a life-changing experience for the right person. Our students have been very successful upon graduation. However, not everyone is a good fit for our group. Few of our PhD students graduate these days since they seem to have significant misconceptions about the requirements for a PhD. Please carefully consider the things mentioned above before applying.

    Prospective applicants should send an email to Dr. Joseph Picone. Please include a current copy of your CV, and examples of some code you have written.

  • (20231005) TUSZ (v2.0.1): We have fixed a few small bugs with the annotations. A total of 35 files were corrected.

  • (20230916) A new version of our annotation tool, nedc_eas, is available here. This is a tool that allows rapid annotation of EEG signals. The tool includes spectrogram and energy plots, and is capable of transcribing data in real time. Learn more about this tool from our IEEE SPMB 2018 paper. This release includes better Windows support and several bug fixes related to montages.

  • (20230819) Summer is almost over, so we celebrated its ending with a group lunch It was nice to see some familiar faces return from their summer internships.

  • (20230731) Summer has gone by quickly. Our three high school summer interns - Nathaniel Gautier, Ean Hodge-Davis and Jose Rivera - have done an excellent job learning Python programming ISIP-style and learning how to run our EEG seizure detection system.

  • (20230527) We have released a real-time EEG seizure detection system based on a ResNet-18 neural network and transfer learning. This package contains a real-time decoder that is described in this publication. This is also part of our real-time demonstration system. The software is available here.

  • (20230425) Congratulations to Somayeh Seifi on passing her PhD prelim exam. Her presentation can be found here.

  • (20230403) A minor update to our EEG seizure detection system (v1.0.1) is is available here. You can learn more about the technology in this publication.

  • (20230330) We were very pleased to welcome back a former student, Emilie Doyle (EE'2016), who returned to participate in a College of Engineering SWE panel discussing career development. Emilie was part of a legendary class loaded with amazing talent. I had the privilege (she probably remembers it differently :) of teaching a group of her colleagues four semesters in a row. She made many great contributions to our lab including this peer mentoring app. She is now pursuing a career as a software engineer at Lockheed Martin.

  • (20230316) Congratulations to an ISIP alumnus, Vince Allen, who recently became a partner at Carstens, Allen & Gourley, LLP. We are very proud of Vince. He walked in my office at MS State many years ago (mid-1990's) and asked me what I thought about electrical engineers becoming a lawyer. I remember asking him if he liked to read a lot (engineers don't usually) and he said yes. I told him he would make a fine lawyer! It has been very exciting to see him build a career in patent law. Becoming a partner is a well-deserved honor for him. I believe this is the only ISIP alumnus that has become a lawyer.

  • (20230310) Congratulations to Nidhi Ram on being a National Honorable Mention winner in the prestigious Award for Aspirations in Computing (AiC) competition. Nidhi, who is a high school student at Lower Moreland High School, has been with ISIP since her April 2021. While in our group she has made several substantial contributions to our research software, and is currently the lead software engineer for the development of the ISIP Machine Learning Demo (IMLD).

  • (20230212) We have released our EEG seizure detection system. The code is available here. You can learn more about the technology in this publication.

2022:

  • (20221207) We have released new versions of the NEDC TUH EEG corpora, and obsoleted our old releases. These new versions contain many enhancements and will be our last releases of this data. See our TUH EEG project for more details.

  • (20220817) ISIP was very honored to host a visit from Dr. Sam McKenzie and Ingrid Lane from the University of New Mexico. Dr. McKenzie is a neuroscientist interested in memory, neural plasticity, how neurons synchronize, and how signals are transformed within and between brain regions. They visited our lab to discuss a joint collaboration that will hopefully lead to the development of a next generation EEG corpus that contains continuous (unpruned) recordings. We are very excited at the chance to collaborate with this team.

  • (20220811) Congratulations to our summer high school students, Micah Jean-Pierre and Zander Hodgson, for completing their summer internship in our lab as Steppingstone Scholars. You can view their final presentation here.

  • (20220807) TUSZ (v1.5.4): We have fixed a few small bugs with the annotations (negative start times) that were introduced due to a problem with our annotation tool. Evaluation results should not change significantly. Please use rsync to update your copy of our distribution.

  • (20220707) NEDC EAS (v5.1.1): This version of our EEG annotation tool includes some bug fixes and some enhanced instructions for installation under Windows.

  • (20220518) TUSZ (v1.5.4): We have rebuilt the dev and eval sets so there is no overlap with the training data. We have also reviewed the annotations.

  • (20220513) Congratulations to our IMLD Team - Thao Cap, Aaron Kreitzer, Matthew Miranda and Dakota Vadimsky - for graduating with their BS degrees. This senior design team developed a Python version of our machine learning demonstration tool. Thao and Matthew also worked as undergraduate researchers in the Institute for Signal and Information Processing.

  • (20220513) IMLD (v1.8.0): The ISIP Machine Learning Demo (IMLD) is now available as a Python application. To learn more about this exciting tool, go here. To learn more about our machine learning course, please visit the course web site.

  • (20220512) TUSZ (v1.5.3): This version of TUSZ is the same as v1.5.2, except the annotations have been converted to our csv format.

  • (20220501) NEDC EAS (v5.1.0): This version of our EEG annotation tool includes enhanced support for xml and csv files. It integrates the same new libraries that support manipulation of annotations.

  • (20220423) We have released NEDC Eval EEG (v5.0.0). This version fully supports our csv and xml annotation file formats.

  • (20220415) Please join us in congratulating the ISIP Machine Learning Demo (IMLD) team for a successful senior design presentation. Their presentation can be viewed here.

  • (20220401) We have made some minor updates to the csv files in the TUAR Corpus. More details can be found: here.

  • (20220331) We have released an update of TUSZ v1.5.2 Corpus that includes the eval data set. Use our anonymous rsync server to retrieve it, or access the data directly from here.

  • (20220318) We now have an updated version of the ISIP env for the Mac, using: /data/isip/tools/mac_x64/AAREADME.txt. It will give you up to date versions of many tools including Python.

  • (20220311) A preliminary draft of the conference website is now live here. The Call for Papers can be accessed on this page.

  • (20220225) Version 1.7.0 of IMLD is now available here. Please let us know if you encounter any issues or bugs.

  • (20220218) We have recently released an official version of the TUH EEH Artifact Corpus (TUEG). The link to the release can be found here. You can also access the release on the downloads page.

  • (20220211) Some of you are having issues using rsync from Windows in order to access our resources. So, we have uploaded this short video that will demonstrate how to install MobaXterm on Windows and how to download data using the command line.

  • (20220124) We are very happy to announce the second release of the Temple University Hospital Digital Pathology (TUDP)! A significant amount of hard work on behalf of a large team of researchers has made this possible. This release only includes annotated slides for breast tissue, however, we have more data on the way. Please review the README file here for more details. Additionally, you can download the data using anonymous rsync:
    rsync -auxv nedc@www.isip.piconepress.com:data/dpath/tuh_dpath/v2.0.0

  • (20220106) Students Thao Cap, Aaron Kreitzer, Matthew Miranda and Dakota Vadimsky presented their senior design project on Interactive Machine Learning Demonstration (IMLD) this past November. We are very proud of their efforts and we are happy to announce the release of IMLD. For more information on the project, view their senior design presentation. Furthermore, the code is available here, along with a demonstration video.

2021:

  • (20211214) Congratulations to all the authors and presenters who were part of IEEE SPMB 2021! Additionally, we would like to thank everyone who attended the conference this year. If you would like to view any of the presentations, recordings of all of them have been uploaded to our conference web site. Click here to access them.

  • (20211202) This Saturday, December 4th, we are hosting our 11th annual symposium. The event will be held virtually and registration fees have been waived. A number of interesting papers regarding EEG science and signal processing will be presented at the conference this year. Furthermore, we will have two excellent plenary talks: one on quantum computing and one one conversational artificial intelligence. Learn more about the presenters and authors attending the symposium this year by visiting the technical program. We are looking forwards to seeing you at IEEE SPMB 2021! Click here to register.

  • (20211126) Congratulations to the team that gave their senior design presentation earlier this month. These students (Thao Cap, Aaron Kreitzer, Matthew Miranda and Dakota Vadimsky) have been working extremely hard on Interactive Machine Learning Demonstration (IMLD). We are very proud of them and their advisor, Dr. Joseph Picone. If you would like to learn more about their work, click here to download their presentation.

  • (20211113) The 2021 IEEE Signal Processing in Medicine and Biology Symposium is only three weeks away! This year's conference will feature 18 oral presentations and 18 poster presentations. For more information regarding presentation times and topics, visit the technical program, which has been posted online. Consider joining us for the conference on December 4th by registering here.

  • (20211015) Our TUH EEG listserv, which features the latest news about our upcoming releases, finally hit 5,000 subscribers (5,040 exactly) This is a great accomplishment for us, especially when considering the size of the research community in this field. Click here to join our listserv!

  • (20211004) IEEE SPMB 2021 is quickly approaching! With the conference only being two months away, we are excited to announce that the first round of the review process for submissions has been completed. This year our acceptance rate was 50% for papers and 70% for abstracts. We will have 18 oral presentations and 18 poster presentations, as well as two plenary speakers. Authors have been notified of the status of their submissions. Revised submissions are due by October 15th. Consider joining us on December 4th for our annual symposium! Click here for more information.

  • (20210903) We are very happy to announce that we have released v5.0.4 of our EEG annotation tool. Now officially known as the NEDC EEG Annotation Tool (EAS), this release supports both Python 3.9 and a new version of PyQT. Moreover, it also supports CSV and XML file formats, both of which we will be using for all our future releases. To access this release directly click here or check it out along with our other resources through our downloads page.

  • (20210827) We submitted two papers and three abstracts to IEEE SPMB this month. The first paper explores Low Latency Real-Time Seizure Detection Using Transfer Deep Learning. The second paper works on Automating Hyperparameter Optimization for Deep Learning Applications. The abstracts explore our TUH EEG Corpus, the Temple University Digital Pathology Corpus, and a Python-Based Interactive Machine Learning Demonstration.

  • (20210809) The Call for Papers for IEEE SPMB 2021 has officially closed. We have received 35 papers and 16 abstract submissions. Consider joining us on December 4th for our annual symposium.

  • (20210809) Congratulations to Vira Oleksyuk on her successful PhD dissertation defense on July 9th. Vira has worked closely with several members of our lab over the past decade, whilst initially pursuing her BSEE degree at Temple. Her expertise in imaging and machine learning has contributed immensely to our group. We wish her the best of luck in her future endeavors!

  • (20210728) Congratulations to Meysam Golmohammadi for his PhD dissertation acceptance into the Temple University Graduate School.

  • (20210721) v5.0.3 of our NEDC Annotation Tool has been released for download. It is also available from our anonymous rsync server. This release now supports Python 3.9 and the newest version of PyQT.

  • (20210714) Congratulations to Meysam Golmohammadi for successfully defending his PhD dissertation titled Deep Architectures for Spatio-Temporal Sequence Recognition with Applications in Automatic Seizure Detection.

  • (20210617) The newest version of our EEG annotation tool is available as a tar file and from our anonymous rsync server (under nedc_annotator/v5.0.2). This version has been a long time coming and is a significant upgrade from previous versions. It now supports csv and xml annotation file formats, which are the preferred formats according to our user community. We have been using this version to develop all our EEG corpora.

  • (20210615) Vinit Shah's PhD dissertation is now available. It is focused on the development of a state of the art seizure detection system by improving the system's ability to detect endpoints of seizure events.

  • (20210602) The deadline for paper submissions for IEEE SPMB 2021 is set for July 1st. Consider joining us in our annual virtual conference this year. More information can be found here.

  • (20210519) Congratulations to Vinit Shah for successfully defending his PhD dissertation titled Improved Segmentation for Automated Seizure Detection Using Channel-Dependent Posteriors.

  • (20210511) Congratulations to Shmyrde Jean-Paul (IT), Safwanur Rahman, Vincent Tchiong, Julien J. Simons (annotation), Lilly Veloso (data collection) and Lynn Vorwick (system management) on their Spring 2021 graduation! Shmyrde has led the IT team and resolved numerous GPU-related issues since joining this group in June 2018. Safwanur has been in our EEG annotation team since September 2017 and has been imperative to the majority of our Seizure Corpus releases. Julien has been a part of our DPATH team since November 2018 and is one of our leading experts at operating the Leica Aperio scanner. Vincent has been an important member of our EEG annotation team since April 2018, having contributed to most of the TUH EEG Seizure Corpus. Lilly has been with our group since February 2017, assisting in data collection for the EEG project. Lynn who has been with us since January 2020, has been vital in cleaning up the environment and NEDC code used in a plethora of our tools. We wish our students all the best in their future endeavors!

  • (20210414) A recent book titled "Biomedical Signal Processing: Innovation and Applications" has been officially published and can be found here. We also have a chapter included within this book regarding Objective Evaluation Metrics for Automatic Classification of EEG Events.

  • (20210409) Our TUH EEG listserv, which has all the latest news on our upcoming releases, has almost hit 5,000 subscribers! Click here if you are interested in joining this listserv.

  • (20210301) IEEE SPMB 2021 has been scheduled for December 4th 2021, consider joining us at our annual symposium. Refer to the Call For Papers for more details.

  • (20210224) Congratulations to Domenic Ochal, a recent 2020 graduate and NEDC alumni who was a member of the EEG annotation team, for being accepted into the Mayo Clinic School of Medicine!

  • (20210218) The Proceedings of the 2020 IEEE Signal Processing in Medicine and Biology Symposium (SPMB) are now available in IEEE Xplore. There were 18 full papers, 17 posters and 2 plenary talks in total. Videos are available from the technical program on the conference web site.

  • (20200218) Congratulations to Ammar Ahmed for successfully defending his PhD dissertation titled Strategies for Radar Communication Spectrum Sharing. Ammar has been a significant contributor to IEEE SPMB. His advisor was Dr. Daniel Zhang.

  • (20210214) The newest version of our EEG annotation tool is available as a tar file and from our anonymous rsync server. This version has been upgraded to run under Python v3.7.x and is much easier to install and run. We have been using this version for over a year to develop v1.5.3 of the TUH EEG Seizure Corpus.

  • (20210107) We have released v2.0.0 of the TUH EEG Artifact Corpus in which the entire signal in each EDF file has been annotated for artifacts. A total of 310 files and 135,863 events have been annotated. This corpus is useful for the evaluation of artifact detection/suppression technology.

  • (20210107) Windows users can look at this brief tutorial on utilizing rsync, which is now included in MobaXterm, to download large corpora.

2020:

  • (20201220) We have released a Python version of our internally-developed software to read an EDF file header and signal data - nedc_pyprint_edf. This software accurately handles all of the EDF files we have seen in our corpora and those provided by collaborators. The software is available from our downloads site.

  • (20201208) Congratulations to our three poster authors - Ahmad Hamid, Safwanur Rahman and Nabila Shawki - for their excellent presentations at IEEE SPMB 2020. More details on the poster presentations can be found here.

  • (20201207) Vinit Shah, our senior PhD student at the Neural Engineering Data Consortium, presented an analysis of our open source EEG scoring software at IEEE SPMB 2020. This software represents a huge step forward for the bioengineering community because it standardizes the scoring process for sequential decoding systems. His presentation can be found here.

  • (20201206) It was exciting to see a former ISIP graduate, Tao Ma, return to give a plenary talk at IEEE SPMB 2020. Tao was the last PhD we graduated from Mississippi State University (ISIP relocated to Temple University in 2009). His talk titled How Has Deep Learning Revolutionized Human Language Technology was an excellent overview of many recent advances in the field. Tao has come a long way from the days when the most exciting part of his day was getting barbecue at Little Dooey's.

  • (20201205) After a year of planning, we successfully wrapped up the 10th anniversary edition of IEEE SPMB. We had 122 participants this year, which was much larger than previous years. We hosted a special session on the Neureka™ 2020 Epilepsy Challenge. A complete record of the conference, including videos of the talks, is available here.

  • (20201204) Join us tomorrow for the 10th annual IEEE Signal Processing in Medicine and Biology Symposium (IEEE SPMB) by registering here.

  • (20201125) IEEE SPMB 2020 is coming up soon! Click here to register for our virtual conference on December 5th.

  • (20201021) This past weekend, a team of our students attended OwlHacks, Temple University's annual hackathon. A variety of workshops and panels were available, including seminars on game development, full-stack web applications and even mobile app development. Our students were able to network with companies such as RSM, Vanguard and Lincoln Financial Group. The hackathon was a thoroughly enjoyable experience for our students and we look forward to future events!

  • (20200930) Congratulations to Tameem Ahsan and Vineetha Mathew, NEDC alumni who were part of our EEG annotation team, on receiving their white coats and becoming part of the class of 2020 at Tufts University School of Medicine. We hope that they will be the first from our lab to become physicians!

  • (20200925) The first round of paper reviews for IEEE SPMB 2020 have been completed. There will be 18 full papers and 12 posters in the technical program. Revised submissions are due by October 15th.

  • (20200925) We have released v4.0.0 of our NEDC annotator with full Python 3 support as well as numerous bug fixes.

  • (20200825) We would like to welcome back our students from their as well as wish everyone good health during these challenging times. More about one of the internships our students undertook can be found here.

  • (20200821) We have released v4.0.0 of our scoring software, NEDC Eval EEG. It is also available on our rsync server.

  • (20200805) Congratulations to Vinit Shah on a successful presentation of his PhD dissertation proposal. Vinit's research focuses on improving segmentation for automated seizure detection.

  • (20200725) We have released an update of the TU Artifact Corpus (TUAR) which can be found here, and is also available from our anonymous rysnc server.

  • (20200714) Our rysnc server is back online and fully operational.

  • (20200702) The IEEE SPMB 2020 call for papers has closed as of July 1st 2020. We plan on accepting 18 papers for oral presentation.

  • (20200701) A search bar has returned to the ISIP web site.

  • (20200624) We have released a video developed for the National Science Foundation (NSF) that features several ISIP students and promotes STEM education.

  • (20200611) In celebration of 10 years of IEEE SPMB, we encourage you to read about the history and original vision of the conference.

  • (20200531) The deadline for paper submission for IEEE SPMB 2020 has been pushed to July 1. Please consider joining us for this special anniversary version of our conference.

  • (20200527) The winners of the Neureka™ 2020 Epilepsy Challenge have just recently been announced here on their website. The competition used a robust version of our database (v1.5.0) to challenge participants. The first place prize was $5000, followed by a second place prize of $2000.

  • (20200514) Congratulations to Tarek Elseify (web, deep learning) and Domenic Ochal, Vineetha Mathew, Tameem Ahsan and Rachel Seo (annotation) for their graduation as part of the class of 2020.

  • (20200513) Amrita Sahu (IEEE SPMB committee member, Temple alumnus) had her hyperspectral imaging work at Altria featured in Vision Systems Design magazine. The project consisted of using hyperspectral data and machine learning algorithms to verify different tobacco grades. Sahu was awarded the "Innovator of the Year" Award within Altria. Congratulations Amrita!

  • (20200511) We have released a demonstration of our real-time seizure detection technology developed with Biosignal Analytics, Inc. This automated seizure detection tool utilizes deep learning technology that has the potential to revolutionize health care in critical settings by allowing physicians to assess patients more safely and efficiently.

  • (20200413) We are delighted to announce an Education Partnership Agreement (EPA) between the Institute for Signal and Information Processing and the Air Force Research Laboratories (AFRL). This partnership will allow ISIP to gain insight into AFRL's computational and technology systems while promoting educational experiences for future engineers and scientists. Some potential areas of collaborative research include deep learning models, intelligent information systems and statistical signal array processing.

  • (20200402) As part of the 10th anniversary of IEEE SPMB, we are collaborating with Novela Neurotech and NeuroTechX to host the Neureka™ 2020 Epilepsy Challenge. We hope you will participate in this very interesting seizure detection challenge.

  • (20200329) We are pleased to announce that a book containing expanded papers from IEEE SPMB 2018 is now available here. A second book covering the 2019 symposium is under the development. Further, the 2020 symposium is now open for submissions. See IEEE SPMB 2020 for more details.

  • (20200125) Katherine Degerberg visited Temple on January 24, 2020 to present a seminar on engineering career management. We had an excellent turnout and the students really enjoyed the event. Relive some fond memories of that event here.

  • (20200124) Please join us in giving Katherine Degerberg a warm welcome to Temple University as she presents a seminar on career development issues in engineering.

2019:

  • (20191216) Congratulations to Meysam Golmohammadi on a successful presentation of his PhD dissertation proposal. Meysam's research focuses on integration of temporal and spatial context into deep learning systems. More details can be found here.

  • (20191210) Congratulations to Nabila Shawki for completing her PhD preliminary exam. She presented very interesting work on the use of autoencoders for feature extraction. We look forward to seeing her complete a dissertation on this topic in a couple of years.

  • (20191215) It is never too early to start planning for IEEE SPMB 2020. The paper submission deadline is June 1, 2020. More information can be found here. IEEE SPMB 2020 will be the 10th year of the symposium. Several special events are planned to commemorate a decade of wonderful experiences.

  • (20191207) IEEE SPMB 2019 was the 9th year for the conference and the 6th year we hosted it at Temple University. 18 papers and 11 posters were presented with an attendance of 62 people. The technical program can be found here.

  • (20191106) IEEE SPMB 2019 is taking place tomorrow at Temple University! Hope to see you there.

  • (20191104) IEEE SPMB 2019 is quickly approaching! Please remember to register to be able to attend SPMB on December 7.

  • (20191025) The TUH EEG Corpus: Annotation File Formats We have released an official document containing information regarding how the annotations are represented in label files.

  • (20190714) Our paper On the Use of Non-Experts for Generation of High-Quality Annotations of Seizure Events has been submitted for publication and is now available.

  • (20190510) The semester is finally over! The ISIP group had our annual end of the academic year party.

  • (20190509) Congratulations to Noah Capp and the class of 2019 on their graduation We wish the best of luck to them.

  • (20190329) Our ECE 1111 students challenged the upper-level ECE 3822 students in an advanced programming competition and came out on top. Fortunately, the ECE 3822 students were not Bamboozled!. Congratulations to all the students for participating in a four-hour Hackathon based on HackerRank problems.

  • (20190323) NEDC TUH EEG Seizure (v1.5.0): This release includes the expansion of the training dataset from 1,984 files to 4,597. Calibration sequences of the new data have been manually annotated and added to the seizure spreadsheet. Annotation corrections were made to the files already existing in the training set.

  • (20190308) IBM TUSZ Pre-Processed Data (v1.0.0): This is our first release of IBMPPD which preprocesses the TUH Seizure Detection Corpus using two methods, both of which use an FFT sliding window approach (STFT) in the beginning.

2018:

  • (20181217) Meysam Golmohammadi attended the International Conference on Machine Learning and Applications (ICMLA) and presented his paper on Deep Architectures for Spatio-Temporal Modeling.

  • (20181206) NEDC TUH EEG Artifact Corpus (v1.0.0): This is our first release of the TUH EEG Artifact Corpus. This corpus was developed to aid in EEG event classification such as seizure detection algorithms. This corpus is a subset of the TUH EEG Corpus and contains files with 5 different types of artifacts: (1) eye movements (EYEM), (2) chewing (CHEW), (3) shivering (SHIV), (4) electrode pop, electrode static, and lead artifacts (ELPP), and (5) muscle artifacts (MUSC).

  • (20181201) Congratulations to our team for hosting IEEE SPMB 2018. We had a lot of wonderful papers and posters presented, as well as two excellent plenary talks. The technical program can be found here.

  • (20181113) The Neural Engineering Data Consortium gave a presentation at the John Hopkins University HLT CoE on EEG Event Classification Using Deep Learning.

  • (20181102) NEDC Eval EEG (v1.3.0): In this release, the FPR definition of the TAES metric has been updated to the standard definition which is #FP / (#FP + #TN) or in other words (1 - TNR).

  • (20181025) The Temple University Hospital Seizure Detection Corpus manuscript has been approved for production and accepted for publication in Frontiers in Neuroinformatics.

  • (20181022) NEDC TUH EEG Seizure (v1.4.0): This release includes improvements to the quality of annotations. Annotation corrections were made in the development test and training sets

  • (20180817) NEDC TUH EEG Seizure (v1.3.0): This release contains quality improvements of the annotations, as manually labeled calibration sequences. The main reason for this release is that we have created a blind evaluation set, often referred to as a held-out set. This is not being released, but will be used in an upcoming Kaggle-style challenge hosted by IBM. More details about this challenge will follow within the next few months.

  • (20180710) NEDC EEG AutoAnnotations (v1.1.0): This release includes the addition of automatically generated annotations using a six-way classification approach. In six-way classification, the first three events are of clinical interest: (1) spike and/or sharp waves (SPSW), (2) periodic lateralized epileptiform discharges (PLED), and (3) generalized periodic epileptiform discharges (GPED). The remaining three events are used to model background noise: (1) eye movement (EYEM), (2) artifacts (ARTF), and (3) background (BCKG).

  • (20180621) The TUH EEG Events Corpus (v1.0.1): The EDF files in the previous release with corrupted headers have been fixed. All files should pass processing using standard open source tools.

  • (20180609) The TUH EEG Corpus (v1.1.0): These annotations are only for EEG files that are under an hour in duration and contain 22 channels.

  • (20180530) The sixth floor lab had a going-away party to celebrate Eva Von Weltin's last week working with us. We are incredibly grateful for having Eva work with us these past few years, and wish her the best in her future endeavors!

  • (20180514) IEEE SPMB 2018 is quickly approaching. Please make sure to submit your papers before the August 1st deadline!

  • (20180509) The end of the semester is finally here! The ISIP group had our annual end of the year party to celebrate the end of the academic year. Time to have a productive summer!

  • (20180506) Congratulations to James Riley McHugh for receiving his B.S. in Electrical Engineering! We are thankful for all the work he's accomplished, and wish him the best in his future endeavors.

  • (20180504) Congratulations to our research team for the acceptance of their abstract submissions for the 2018 ISMB conference! Abstracts discuss the automatic interpretation of abnormal EEGs, curriculum based deep learning techniques, and a deep residual learning framework for convolutional neural networks.

  • (20180501) NEDC Eval EEG (v1.2.0): This scoring software results using 5 different scoring methods: NIST, DP Align, Epoch Based, Any-Overlap, and TAES.

  • (20180417) The TUH EEG Corpus (v1.1.0): This is the official version of the TUH EEG Corpus. This release organizes each session by the montage definition it fits. File naming conventions have been adjusted and files not containing any usable brain signal have been removed.

  • (20180416) The TUH EEG Seizure Corpus (v1.2.1): This release includes small annotation fixes and an addition to the _SEIZURES spreadsheet of calibration start and end times, as well as other small cleanups.

  • (20180416) The TUH EEG Seizure Corpus (v1.2.1) This release includes small annotation fixes and an addition to the _SEIZURES spreadsheet of calibration start and end times.

  • (20180412) The TUH Abnormal EEG Corpus (v2.0.0): This is a subset of the TUH EEG corpus that can be used for automatic detection of abnormal EEGs. This release contains patient numbers that have been re-mapped to be consistent with v1.1.0 of TUH EEG. Some cleanups including the removal of duplicate files have been made.

  • (20180412) The TUH EEG Slowing Corpus (v1.0.1): This is a small data set that can be used to study the difference between slowing and seizures in EEGs. This release contains patient numbers that have been re-mapped to be consistent with v1.1.0 of TUH EEG.

  • (20180327) The TUH EEG Six-Way Event Classification Corpus (v1.0.0): This release contains the data used to develop our initial version of AutoEEG. Sections of EEG signals are annotated for one of 6 events: spike, gped, pled, eye movement, artifact and background. Please review the AAREADME file for more information about the release.

  • (20180208) Visualize EEG signals with our Python-based visualization tool! Functionality includes an integrated annotation system (which can create and review label files), configurable EDF scrolling, visualization in the form of spectrograms, and much more! The tool has been developed to assist with the annotation of TUH EEG and to support our machine learning research. For more information about the tool, click here.

  • (20180108) Happy New Years! We have big events planned for 2018, and we are excited to seeing you all at IEEE SPMB 2018 this year!

2017:

  • (20170923) v1.0.0 of the TUH EEG Corpus was finally released. See TUH EEG Corpus for more details.

  • (20170920) v1.0.0 of the TUH EEG Slowing Corpus has been released. See TUH EEG Slowing Corpus for more details.

  • (20170805) v1.1.1 of the TUH EEG Seizure Corpus has been released. See TUH EEG Seizure Corpus for more details.

  • (20170727) Congratulations to our GETS summer students for successful completion of their 5-week summer course on machine learning.

  • (20170722) v1.1.0 of the TUH EEG Seizure Corpus has been released. See TUH EEG Seizure Corpus for more details.

  • (20170629) Today, the research team celebrated the departure of two significant graduates, Silvia Lopez and Aaron Gross, over lunch. Prior to lunch, Silvia gave an outstanding presentation on her thesis proposal, found here.

  • (20170621) The research team has submitted their paper in the NIPS2017 conference titled "Deep Architectures for Automated Seizure Detection in Scalp EEGs."

  • (20170621) On behalf of a very hard-working group of undergraduates, we are pleased to announce the release of v1.0.4 of the TUH EEG Seizure Corpus.

  • (20170601) The 7th annual IEEE Signal Processing in Medicine and Biology Symposium (IEEE SPMB 2017) will be held on December 2, 2017 at Temple University in Philadelphia. Please refer to the conference website and the Call For Papers for more details.

  • (20170523) Congratulations to our newest 2017 Spring graduates, Elliott Krome and Aaron Gross, for their excellent achievements and progress made for the NEDC EEG Research group.

  • (20170428) Today, the research team presented the demo at Temple's Innovation Showcase! The presentation materials may be found here.

  • (20170427) Congratulations to Vinit Shah for passing his PhD Preliminary Exam! A video of his presentation and more information can be found here.

  • (20170426) The manually annotated TUH EEG Seizure Detection Corpus has finally been released! To learn more about this database or to download it, click here.

  • (20170407) Drs. Picone and Obeid will be publishing a book chapter titled "Machine Learning Approaches to Automatic Interpretation of EEGs" soon. This will appear as a chapter in the CRC Press book "Biomedical Signal Processing in Big Data" which is expected to be published in Fall 2017. A prerelease of the chapter is available here. This chapter presents a concise overview of the application of machine learning and big data to EEG interpretation.

  • (20170314) Seyedeh Saeedeh, Meysam Golmohammadi, and Dawer Jamshed have attended the annual Columbia Workshop to learn more about Brain Circuits, Memory, and Computation. Click here to learn more.

  • (20170209) Meysam Golmohammadi presented a poster on the TUH EEG Seizure Detection Corpus at the annual meeting of the American Clinical Neurophysiology Society. His poster can be viewed here.

2016:

  • (20161221) The fall semester has come to an end and we have celebrated the holiday festivities with a Christmas party! The photos of the party can be found here.

  • (20161214) Congratulations to Amir Harati, who has successfully presented a paper titled "A nonparametric Bayesian approach for automatic discovery of a lexicon and acoustic units" at IEEE Workshop SLT 2016. You can view his work here.

  • (20161204) Congratulations to our team for hosting IEEE SPMB 2016. We had over 70 attendees, 24 papers and 9 posters presented, and two excellent plenary talks. The technical program can be found here.

  • (20161130) We attended the NIH BD2K All Hands Meeting to present our latest work on cohort retrieval. Our presentations included a demo of AutoEEG with a new cohort retrieval feature.

  • (20160930) It is pizza time again in ISIP. This doesn't happen too often but is one of those can't miss events -- homemade pizza!

  • (20160916) Congratulations again to Amir Harati, who had his paper titled A Nonparametric Bayesian Approach for Automatic Discovery of a Lexicon and Acoustic Units accepted to the 2016 IEEE Workshop on Spoken Language Technology.

  • (20160912) Congratulations to Amir Harati, who has successfully presented a portion of his dissertation titled "A Nonparametric Bayesian Approach For Spoken Term Detection by Example Query" at INTERSPEECH 2016. You can view his work here.

  • (20160825) The Fall 2016 semester is about to start, so it is time for our beginning of semester lunch to welcome our new students. Two graduate students are joining our group, and we expect to add two undergraduates as well. Unfortunately, none of them could attend their welcoming party :( but that didn't stop us from enjoying a nice lunch.

  • (20160729) It is the end of the summer and it is time to say goodbye to our Brazilian exchange student, Matheus Guedes De Andrade. He had a very productive summer working on our seizure detection project. His poster presentation can be found here.

  • (20160715) Congratulations to our Summer Of Code undergraduates, who presented at this year's College of Engineering Research Day. You can view their posters here. Enjoy some pictures from the conference.

  • (20160506) Congratulations to our Spring 2016 graduates. Amir finally walked! We will miss Sammy and Devin. Enjoy some pictures from our graduation party.

  • (20160428) With much sadness we wish Sammy Deegan and Devin Trejo a fond farewell and send them off to exciting new engineering careers. Both are graduating this semester and will leave Temple University. They each made enormous contributions to our group and will be missed. Enjoy some pictures from their going-away party.

  • (20160324) Scott Yang joins our group as our first postdoc. Scott will lead our NIH-funded project titled "Automatic discovery and processing of EEG cohorts from clinical records". Go here to learn more about this project.

  • (20160225) Congratulations to Saeedeh Khoshgoftar for passing her PhD Preliminary Exam today. His paper and presentation are located here.

  • (20160223) Congratulations to Meysam Golmohammadi for passing his PhD Preliminary Exam today. His paper and presentation are located here.

  • (20160215) Congratulations to Amir Harati for his recent journal publication titled "A Doubly Hierarchical Dirichlet Process Hidden Markov Model with a Non-Ergodic Structure." The hardcopy version of the IEEE/ACM Transactions on Audio, Speech, and Language Processing containing his paper was released this month.

2015:

  • (20151112) We attended the NIH BD2K All-Hands Meeting to present several posters on our project titled "Automatic discovery and processing of EEG cohorts from clinical records". To learn more about this research, go here.

  • (20150729) The 2015 version of our infamous Summer of Code is rapidly coming to an end, which means it is time for a party. Enjoy some pictures from our going away party. To learn more about the research conducted this summer, go here.

  • (20150724) Congratulations to Devin Trejo and Fabricio Teles Dutra Goncalves for presenting posters at the College of Engineering Undergraduate Summer Research Workshop. Some pictures from this event can be found here.

  • (20150714) Meysam Golmohammadi, Iyad Obeid and Lou Bucelli completed the first phase of I-Corps training in Chicago.

  • (20150528) Congratulations to Dr. Steven Tobochnik, who graduated from the Temple University School of Medicine in May, and is on his way to a residency at Columbia University New York Presbyterian Hospital. Steven was instrumental in creating our AutoEEG technology by teaching us how to read EEGs and hand-labeling enough data to build good models. We hope to continue collaborating with him once he settles into his new home in NYC. We will miss him. He was an excellent teacher and collaborator.

  • (20150508) Congratulations to Silvia Lopez who graduated with her BS in Electrical Engineering. Silvia will start her MS degree in Fall 2015 and continue to play a prominent role in our EEG projects.

  • (20150424) We participated in the kickoff meeting for Temple Ventures by demonstrating our AutoEEG technology. Learn more about our project here.

  • (20150402) Join us in congratulating a colleague, Kazuhiro Kondo, on his promotion to full professor at Yamagata University in Japan. We first met Kondo-sensei in the early 1991 at Texas Instruments in Japan. We have been friends ever since. He has been highly productive was a professor and is author of the book Subjective Quality Measurement of Speech: Its Evaluation, Estimation and Applications (Signals and Communication Technology).

  • (20150226) Several students presented posters at the 2015 Temple University College of Engineering Research Poster Competition during Engineer's Week. See Temple CoE Posters to view the work that was presented..

2014:

  • (20141213) We hosted the 2014 IEEE Signal Processing in Medicine and Biology Symposium. The conference web site is located here.

  • (20140815) Our long-time colleague, Dr. Georgios Lazarou, returned to academia this fall when he joined the Department of Electrical and Computer Engineering at University of South Alabama in Mobile, Alabama. He was previously living in New York City where he was instrumental in reengineering the communications systems for the New York City Transit system. We wish him well in his new endeavor and look forward to our continued collaborations.

  • (20140802) Silvia Lopez presents a poster titled "Automatic Interpretation of EEGs Using Big Data" at the 2014 Temple College of Engineering Research Experience for Undergraduate Students Conference.

  • (20140326) Amir Harati presented the most complete version of his dissertation work to date at the 48th Annual Conference on Information Sciences and Systems. Learn more about his groundbreaking work on nonparametric Bayesian models here.

2013:

  • (20140102) In 2013, the speech research community lost one of its beloved members, David Pallett. For many years, Dave guided the Speech Group at the National Institute for Standards and Technology (NIST), and transformed speech research from an unprincipled discipline to a science. Human language technology conferences just aren't the same without Dave's commanding presence showing an updated common evaluations slide.

  • (20131207) Joe Picone presented a poster on the TUH EEG Corpus at the IEEE Signal Processing in Medicine and Biology Symposium (SPMB13). Temple University will host this conference in 2014. Learn more about our EEG work here.

  • (20131206) John Steinberg presented his MS thesis work at the International Conference on Systems, Computing Sciences and Software Engineering (SCSS 13). You can read his paper here.

  • (20131204) Our first paper on the TUH EEG database was presented at the first IEEE Global Conference on Signal and Information Processing ( GlobalSIP 2013). We also organized a symposium titled "Advancing Neural Engineering Through Big Data." More information about the symposium and our paper can be found here.

  • (20130831) Amir Harati's paper on his dissertation work was presented at INTERSPEECH 2013. Enjoy some memories from our trip to Lyon, France. His paper was well-received and people are looking forward to his recognition results on conversational speech.

  • (20130812) Shuang Lu presented a conference paper on her dissertation work on fingerspelling at IPCV'13 in Las Vegas, Nevada, USA. Despite being known as Sin City, The City That Never Sleeps, and the Gambling Capital of the World, Shuang only was willing to share this picture showing her hard at work. I guess what happens in Vegas stays in Vegas :)

  • (20130527) Enjoy some pictures from John Steinberg's going-away party. It was held at the Dock Street Brewing Company in West Philadelphia. Good luck to John as he heads to the SCALE summer workshop.

  • (20130516) John Steinberg graduated with his MS. His thesis work can be found here. Ilyana Mushaeva graduated with here MS also. Sung Cho and Vira Oleksyuk graduated with their BS degrees. Enjoy some moments from Temple's 126th graduate ceremony here. All have played an instrumental role in getting ISIP established at Temple and contributed significantly to the department in various administrative capacities.

  • (20130418) Amir Harati passed his PhD proposal defense today. He is the first ISIP student at Temple University to pass this milestone. View his presentation here. Congratulations to Amir!

2012:

  • (20120222) View the papers we presented at Temple's College of Engineering Research Day here.

  • (20130216) View our CogSIMA conference poster on keyword search here. There is also a demo of the system available at Keyword Search Demo. This presentation is the first detailed look at predicting the quality of a search term and what factors contribute to poor keyword search performance.

  • (20130101) View our ICASSP 2013 submission here. This is the basis of Amir Harati's dissertation work on nonparametric Bayesian models for speech recognition.

  • (20121220) View our paper on keyword search term reliability to be presented at the 2013 IEEE International Multi-Disciplinary Conference on Cognitive Methods in Situation Awareness and Decision Support.

  • (20121215) This figure represents a nice overview of ASR learning paradigms. It appeared in a paper by Jim Glass and his colleagues at MIT. See unsupervised learning for more details.

  • (20120301) Enjoy our demonstration of keyword search term strength.