Search Constraints
Number of results to display per page
View results as:
Search Results
-
- Creator:
- Ali, Hashim, Subramani, Surya , Sudhir, Shefali , Varahamurthy, Raksha , and Malik, Hafiz
- Description:
- Voice-cloning (VC) systems have seen an exceptional increase in the realism of synthesized speech in recent years. The high quality of synthesized speech and the availability of low-cost VC services have given rise to many potential abuses of this technology such as online smearing campaigns and dissemination of fabricated information etc. A number of detection methodologies have been proposed over the years that can detect voice spoofs with reasonably good accuracy. However, these methodologies are mostly evaluated on clean audio databases, such as Asvspoof 2019. This research aims to evaluate state-of-the-art (SOTA) Audio Spoof Detection approaches in the presence of laundering attacks. In that regard, a new laundering attack database, called ASVspoof Laundering Database, is created. This database is based on the ASVspoof 2019 LA eval database comprising a total of 1388.22 hours of audio recordings. Seven SOTA audio spoof detection approaches are evaluated on this laundered database. The results indicate that SOTA systems perform poorly in the presence of aggressive laundering attacks, especially reverberation and additive noise attacks. This suggests the need for robust audio spoof detection.
- Keyword:
- Audio Forensics, Audio Antispoofing, Audio Deepfakes, ASVSpoof, and Machine Learning
- Discipline:
- Engineering
-
- Creator:
- Asthana, Vishwaratn, Monovoukas, Demetri, Kucharski, Kevin, Chopra, Zoey, Perkins, Sidney, and Bugga, Pallavi
- Description:
- It is difficult to model outcomes in patients post-ST Elevation Myocardial Infarction (STEMI), particularly onset of heart failure. The acute insult following a myocardial infarction and chronic degeneration seen in HF involve a similar process where a loss of cardiomyocytes and abnormal remodeling lead to pump failure. This process may alter the strength and direction of the heart’s net depolarization signal. The investigators hypothesized that changes over time in unique parameters extracted using vectorcardiography (VCG) have the potential to predict clinical outcomes in patients post-STEMI and could eventually be used as a non-invasive and cost-effective surveillance tool for characterizing the severity and progression of HF to guide evidence-based therapies.
- Keyword:
- Cardiology, ECG, VCG, STEMI, and Heart Failure
- Discipline:
- Health Sciences
-
- Creator:
- Fu, Xun, Zhang, Bohao, Weber, Ceri J., Cooper, Kimberly L., Vasudevan, Ram, and Moore, Talia Y.
- Description:
- Tails used as inertial appendages induce body rotations of animals and robots---a phenomenon that is governed largely by the ratio of the body and tail moments of inertia. However, vertebrate tails have more degrees of freedom (e.g., number of joints, rotational axes) than most current theoretical models and robotic tails. To understand how morphology affects inertial appendage function, we developed an optimization-based approach that finds the maximally effective tail trajectory and measures error from a target trajectory. For tails of equal total length and mass, increasing the number of equal-length joints increased the complexity of maximally effective tail motions. When we optimized the relative lengths of tail bones while keeping the total tail length, mass, and number of joints the same, this optimization-based approach found that the lengths match the pattern found in the tail bones of mammals specialized for inertial maneuvering. In both experiments, adding joints enhanced the performance of the inertial appendage, but with diminishing returns, largely due to the total control effort constraint. This optimization-based simulation can compare the maximum performance of diverse inertial appendages that dynamically vary in moment of inertia in 3D space, predict inertial capabilities from skeletal data, and inform the design of robotic inertial appendages.
- Keyword:
- simulation, inertial maneuvering, caudal vertebrae, trajectory optimization, and reconfigurable appendages
- Citation to related publication:
- Xun Fu, Bohao Zhang, Ceri J. Weber, Kimberly L. Cooper, Ram Vasudevan, Talia Y. Moore. (in review) Jointed tails enhance control of three-dimensional body rotation.
- Discipline:
- Engineering and Science
-
- Creator:
- Gottesman, Ari
- Description:
- This study analyzes correlations between magnetic field data from closely-spaced pairs of ground magnetometers to observe the spatial scale of ionospheric current signatures. Correlations were mainly calculated in 7.5 minute intervals for periods of multiple days. Distributions were taken from the collection of these 7.5 minute intervals to identify the amount of time where the magnetometers were observing "similar" or "different" ionospheric signatures. The raw magnetometer data was taken from two geomagnetic storms: one taking place on 7-8 September, 2017, and the other taking place on 23-24 March, 2023. These periods were selected due to the presence of both high and low geomagnetic activity. The final distributions calculated from this analysis are available in Correlation_Distributions.csv.
- Keyword:
- Space Weather Impacts, Geomagnetically Induced Currents, and GIC
- Discipline:
- Science
-
- Creator:
- Zhang, Qiaoning
- Description:
- The widespread acceptance of healthcare robot at home is hindered by a lack of clarity regarding optimal design features, particularly among users with varying levels of knowledge and attitudes towards this emerging technology. To address this, this study applies the Kano model to systematically identify and prioritize the features of healthcare robots, factoring in older adults diverse robot-related knowledge and attitudes towards robots.
- Discipline:
- Social Sciences
-
- Creator:
- Irani, Sanaya , Tolia, Sangini, Finks, Jonathan, and Sandhu, Gurjit
- Description:
- Program Description DoT was founded in 2012 with a mission to increase diversity amongst medical professionals by preparing students from underrepresented communities in Detroit to successfully pursue careers in healthcare. Our program builds on a partnership between Cass Technical High School (CTHS) and the University of Michigan Medical School (UMMS). The CTHS student body is reflective of the Detroit population with more than 80% of students identifying with racial and ethnic minority backgrounds. Students with an interest in healthcare apply for the program as ninth graders. In recent years, the program has received over 60 applications for approximately 30 positions in each grade. DoT’s unique strength lies in its longitudinal structure. There are three branches of the program – Foundations (ninth and tenth grade), Rising (eleventh and twelfth grade) and Succeed (undergraduate). Ninth graders start out in DoT Foundations. Each student is paired with a first-year medical student mentor at UMMS for the entire academic year. DoT students travel to UMMS every month for a visit day, with activities designed to give students hands-on experiences in healthcare, such as suturing and ultrasound skills in the simulation center, and clinical shadowing. Students then meet with their medical student mentor over lunch. The latter part of the day is dedicated to working on their capstone projects. For the capstone projects, students work in small teams led by medical student leaders to identify a community health issue, partner with a local organization, and present their proposed solutions at a formal symposium at the end of the year. , Transition to Virtual Programming In light of the recent COVID-19 pandemic, a growing number of universities cancelled all campus events including those of pipeline programs. We felt that our programming offered an important service to our students that would be greatly missed, so our team worked to quickly create and implement a virtual program. We ensured that each of our students had access to technology at home and those who did not were offered scholarships. During our introductory student session and new parent meeting, our leadership team discussed how to set up a Gmail email address for weekly communications and taught the students how to use Zoom, Google Drive, Google Docs and Google Sheets for online learning collaboration. For the virtual Foundations program, we offered 1-hour seminars each month, where a physician was invited to give a 30-minute presentation about different organ systems, followed by a 30-minute case-based session where students worked with medical student mentors to apply their new knowledge. We also created novel sessions such as “The Path to College and Medical School” and collaborated with members of the Black Medical Association (BMA) and Latin American and Native American Medical Association (LANAMA) to host a panel session where students could learn from medical students who identified as URiM. For the mentorship aspect, we created “pods” of Foundations, Rising, and Succeed students along with medical student and physician mentors. The Foundations students and mentors met every month for an hour on Zoom, a virtual communication platform, to work on their Capstone project. Rising and Succeed students joined the group for three full-pod meetings. The goal was to increase near-peer mentorship and connections between DoT students at all levels. , and Study Population Due to the virtual nature of the 2020-2021 program, we accepted 100% of 9th grade applicants from CTHS. We also expanded our reach to a new school, The School at Marygrove (TSM), which is also located in Detroit, Michigan. TSM is involved in the Detroit-20 Partnership with the University of Michigan College of Education and includes a novel three-year residency program for novice teachers. During the 2020-2021 school year, 108 students participated in the Foundations programming with 72 of them being 9th graders and 36 being 10th graders. The students were mostly from CTHS with 12 students out of the 108 total being from TSM. Students were predominantly from an African American/Black racial background (68.4% from N=98 respondents). The students were representative of their respective schools. The majority of students at CTHS identify as black, come from low-income homes, and have variable levels of parental education.
- Keyword:
- pipeline program, Underrepresented in medicine, Mentorship, Medical education, and COVID-19
- Discipline:
- Health Sciences
-
- Creator:
- University of Michigan Museum of Zoology
- Description:
- Scan of specimen ummz:mammals:97068 (Ammotragus lervia) - Skull. Raw Dataset includes 1601 TIF images (each 1972 x 1578 x 1 voxel at 0.1215074 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction. and Scan of specimen ummz:mammals:97068 (Ammotragus lervia) - Skull. Reconstructed Dataset includes 1972 TIF images (each 1972 x 1578 x 1 voxel at 0.121507 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction.
- Keyword:
- Animalia, Chordata, Mammalia, Artiodactyla, Bovidae, Ammotragus lervia, 1987319019, computed tomography, X-ray, and 3D
- Citation to related publication:
- For more information on the original UMMZ specimen, see: https://www.gbif.org/occurrence/1987319019
- Discipline:
- Science
-
- Creator:
- University of Michigan Museum of Zoology
- Description:
- Scan of specimen ummz:mammals:103305 (Eudorcas THOMSONII) - Skull. Raw Dataset includes 1601 TIF images (each 1870 x 924 x 1 voxel at 0.0912381 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction. and Scan of specimen ummz:mammals:103305 (Eudorcas THOMSONII) - Skull. Reconstructed Dataset includes 1636 TIF images (each 1870 x 924 x 1 voxel at 0.091238 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction.
- Keyword:
- Animalia, Chordata, Mammalia, Artiodactyla, Bovidae, Eudorcas THOMSONII, 1987323462, computed tomography, X-ray, and 3D
- Citation to related publication:
- For more information on the original UMMZ specimen, see: https://www.gbif.org/occurrence/1987323462
- Discipline:
- Science
-
- Creator:
- University of Michigan Museum of Zoology
- Description:
- Scan of specimen ummz:mammals:125684 (Capreolus capreolus) - Skull. Raw Dataset includes 1601 TIF images (each 1555 x 1967 x 1 voxel at 0.1215076 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction. and Scan of specimen ummz:mammals:125684 (Capreolus capreolus) - Skull. Reconstructed Dataset includes 1411 TIF images (each 1555 x 1967 x 1 voxel at 0.121508 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction.
- Keyword:
- Animalia, Chordata, Mammalia, Artiodactyla, Cervidae, Capreolus capreolus, 1987249347, computed tomography, X-ray, and 3D
- Citation to related publication:
- For more information on the original UMMZ specimen, see: https://www.gbif.org/occurrence/1987249347
- Discipline:
- Science
-
- Creator:
- University of Michigan Museum of Zoology
- Description:
- Scan of specimen ummz:mammals:178562 (Ourebia ourebi) - Skull. Raw Dataset includes 1601 TIF images (each 791 x 1671 x 1 voxel at 0.1214979 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction. and Scan of specimen ummz:mammals:178562 (Ourebia ourebi) - Skull. Reconstructed Dataset includes 1446 TIF images (each 791 x 1671 x 1 voxel at 0.121498 mm resolution, derived from 1601 scan projections), xtek and vgi files for volume reconstruction.
- Keyword:
- Animalia, Chordata, Mammalia, Artiodactyla, Bovidae, Ourebia ourebi, 1987263154, computed tomography, X-ray, and 3D
- Citation to related publication:
- For more information on the original UMMZ specimen, see: https://www.gbif.org/occurrence/1987263154
- Discipline:
- Science