Thamidul Islam Tonmoy

Thamidul Islam Tonmoy

Auto-generated profile, based on publicly available information

Thamidul Islam Tonmoy

Department of Bioengineering, University of California, Riverside, USA

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 April 2019 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
UKUday Kamal
Uday Kamal
TTThamidul Islam Tonmoy
Thamidul Islam Tonmoy
SDSowmitra Das
Sowmitra Das
KHKamrul Hasan
Kamrul Hasan
Traffic sign detection is a central part of autonomous vehicle technology. Recent advances in deep learning algorithms have motivated researchers to use neural networks to perform this task. In this paper, we look at traffic sign detection as an image segmentation problem and propose a deep convolutional neural network-based approach to solve it. To this end, we propose a new network, the SegU-Net, which we form by merging the state-of-the-art segmentation architectures-SegNet and U-Net to detect traffic signs from video sequences. For training the network, we use the Tversky loss function constrained by an L1 term instead of the intersection over union loss traditionally used to train segmentation networks. We use a separate network, inspired by the VGG-16 architecture, to classify the detected signs. The networks are trained on the challenge free sequences of the CURE-TSD dataset. Our proposed network outperforms the state-of-the-art object detection networks, such as the Faster R-CNN inception Resnet V2 and R-FCN Resnet 101, by a large margin and obtains a precision and recall of 94.60% and 80.21%, respectively, which is the current state of the art on this part of the dataset. In addition, the network is tested on the German Traffic Sign Detection Benchmark (GTSDB) dataset, where it achieves a precision and recall of 95.29% and 89.01%, respectively. This is on a par with the performance of the aforementioned object detection networks. These results prove the generalizability of the proposed architecture and its suitability for robust traffic sign detection in autonomous vehicles.

17 October 2023 in npj Regenerative Medicine
Verified
Journal Article
Open access
PDF file
YTYouyi Tai
Youyi Tai
TTThamidul Islam Tonmoy
Thamidul Islam Tonmoy
SWShwe Win
Shwe Win
NBNatasha T. Brinkley
Natasha T. Brinkley
BPB. Hyle Park
B. Hyle Park
Jin NamJin Nam
Jin Nam
To address limitations in current approaches for treating large peripheral nerve defects, the presented study evaluated the feasibility of functional material-mediated physical stimuli on peripheral nerve regeneration. Electrospun piezoelectric poly(vinylidene fluoride-trifluoroethylene) nanofibers were utilized to deliver mechanical actuation-activated electrical stimulation to nerve cells/tissues in a non-invasive manner. Using morphologically and piezoelectrically optimized nanofibers for neurite extension and Schwann cell maturation based on in vitro experiments, piezoelectric nerve conduits were synthesized and implanted in a rat sciatic nerve transection model to bridge a critical-sized sciatic nerve defect (15 mm). A therapeutic shockwave system was utilized to periodically activate the piezoelectric effect of the implanted nerve conduit on demand. The piezoelectric nerve conduit-mediated mechano-electrical stimulation (MES) induced enhanced peripheral nerve regeneration, resulting in full axon reconnection with myelin regeneration from the proximal to the distal ends over the critical-sized nerve gap. In comparison, a control group, in which the implanted piezoelectric conduits were not activated in vivo, failed to exhibit such nerve regeneration. In addition, at both proximal and distal ends of the implanted conduits, a decreased number of damaged myelination (ovoids), an increased number of myelinated nerves, and a larger axonal diameter were observed under the MES condition as compared to the control condition. Furthermore, unlike the control group, the MES condition exhibited a superior functional nerve recovery, assessed by walking track analysis and polarization-sensitive optical coherence tomography, demonstrating the significant potential of the piezoelectric conduit-based physical stimulation approach for the treatment of peripheral nerve injury.

30 July 2020 in Neural Computing and Applications
Verified
Journal Article
Closed access
PDF file
ARAbdul Muntakim Rafi
Abdul Muntakim Rafi
TTThamidul Islam Tonmoy
Thamidul Islam Tonmoy
UKUday Kamal
Uday Kamal
JWJonathan Wu
Jonathan Wu
KHKamrul Hasan
Kamrul Hasan

7 December 2022 in Biomolecules
Verified
Journal Article
Open access
PDF file
MSMartina M. Sanchez
Martina M. Sanchez
TTThamidul Islam Tonmoy
Thamidul Islam Tonmoy
BPB. Hyle Park
B. Hyle Park
JMJoshua T. Morgan
Joshua T. Morgan
Photoaging is an important extrinsic aging factor leading to altered skin morphology and reduced function. Prior work has revealed a connection between photoaging and loss of subcutaneous fat. Currently, primary models for studying this are in vivo (human samples or animal models) or in vitro models, including human skin equivalents (HSEs). In vivo models are limited by accessibility and cost, while HSEs typically do not include a subcutaneous adipose component. To address this, we developed an “adipose-vascular” HSE (AVHSE) culture method, which includes both hypodermal adipose and vascular cells. Furthermore, we tested AVHSE as a potential model for hypodermal adipose aging via exposure to 0.45 ± 0.15 mW/cm2 385 nm light (UVA). One week of 2 h daily UVA exposure had limited impact on epidermal and vascular components of the AVHSE, but significantly reduced adiposity by approximately 50%. Overall, we have developed a novel method for generating HSE that include vascular and adipose components and demonstrated potential as an aging model using photoaging as an example.

27 June 2020 in arXiv
Verified
Preprint
Open access
PDF file
ARAbdul Muntakim Rafi
Abdul Muntakim Rafi
TTThamidul Islam Tonmoy
Thamidul Islam Tonmoy
UKUday Kamal
Uday Kamal
JWJonathan Wu
Jonathan Wu
KHKamrul Hasan
Kamrul Hasan
Camera model identification (CMI) has gained significant importance in image forensics as digitally altered images are becoming increasingly commonplace. In this paper, a novel convolutional neural network (CNN) architecture is proposed for CMI with emphasis given on the preprocessing task considered to be inevitable for removing the scene content that heavily obscures the camera model fingerprints. Unlike the conventional approaches where fixed filters are used for preprocessing, the proposed remnant blocks, when coupled with a classification block and trained end-to-end minimizing the classification loss, learn to suppress the unnecessary image contents dynamically. This helps the classification block extract more robust camera model-specific features for CMI from the remnant of the image. The whole network, called RemNet, consisting of a preprocessing block and a shallow classification block, when trained on 18 models from the Dresden database, shows 100% accuracy for 16 camera models with an overall accuracy of 97.59% on test images from unseen devices, outperforming the state of the art deep CNNs used in CMI. Furthermore, the proposed remnant blocks, when cascaded with the existing deep CNNs, e.g., ResNet, DenseNet, boost their performances by a large margin. The proposed approach proves to be very robust in identifying the source camera models, even if the original images are post-processed. It also achieves an overall accuracy of 95.11% on the IEEE Signal Processing Cup 2018 dataset, which indicates its generalizability.