Search

Your search keyword '"Munir, Muhammad"' showing total 3,854 results
Start Over Author "Munir, Muhammad"
3,854 results on '"Munir, Muhammad"'

3. Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment

Author

Danish, Muhammad Sohail, Khan, Muhammad Haris, Munir, Muhammad Akhtar, Sarfraz, M. Saquib, and Ali, Mohsen

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

In this work, we tackle the problem of domain generalization for object detection, specifically focusing on the scenario where only a single source domain is available. We propose an effective approach that involves two key steps: diversifying the source domain and aligning detections based on class prediction confidence and localization. Firstly, we demonstrate that by carefully selecting a set of augmentations, a base detector can outperform existing methods for single domain generalization by a good margin. This highlights the importance of domain diversification in improving the performance of object detectors. Secondly, we introduce a method to align detections from multiple views, considering both classification and localization outputs. This alignment procedure leads to better generalized and well-calibrated object detector models, which are crucial for accurate decision-making in safety-critical applications. Our approach is detector-agnostic and can be seamlessly applied to both single-stage and two-stage detectors. To validate the effectiveness of our proposed methods, we conduct extensive experiments and ablations on challenging domain-shift scenarios. The results consistently demonstrate the superiority of our approach compared to existing methods. Our code and models are available at: https://github.com/msohaildanish/DivAlign

Published
2024

4. Pericardial effusion requiring intervention in patients undergoing leadless pacemaker implantation: A real-world analysis from the National Inpatient Sample database.

Author

Khan, Muhammad, Sattar, Yasar, Alruwaili, Waleed, Nassar, Sameh, Alhajji, Mohamed, Alyami, Bandar, Nguyen, Amanda, Neely, Joseph, Abideen Asad, Zain, Agarwal, Siddharth, Raina, Sameer, Balla, Sudarshan, Nguyen, Bao, Fan, Dali, Darden, Douglas, and Munir, Muhammad

Subjects
Leadless pacemakers, Outcomes, Pericardial effusion, Predictors
Abstract

BACKGROUND: Pericardial effusion requiring percutaneous or surgical-based intervention remains an important complication of a leadless pacemaker implantation. OBJECTIVE: The study sought to determine real-world prevalence, risk factors, and associated outcomes of pericardial effusion requiring intervention in leadless pacemaker implantations. METHODS: The National Inpatient Sample and International Classification of Diseases-Tenth Revision codes were used to identify patients who underwent leadless pacemaker implantations during the years 2016 to 2020. The outcomes assessed in our study included prevalence of pericardial effusion requiring intervention, other procedural complications, and in-hospital outcomes. Predictors of pericardial effusion were also analyzed. RESULTS: Pericardial effusion requiring intervention occurred in a total of 325 (1.1%) leadless pacemaker implantations. Patient-level characteristics that predicted development of a serious pericardial effusion included >75 years of age (odds ratio [OR] 1.38, 95% confidence interval [CI] 1.08-1.75), female sex (OR 2.03, 95% CI 1.62-2.55), coagulopathy (OR 1.50, 95% CI 1.12-1.99), chronic pulmonary disease (OR 1.36, 95% CI 1.07-1.74), chronic kidney disease (OR 1.53, 95% CI 1.22-1.94), and connective tissue disorders (OR 2.98, 95% CI 2.02-4.39). Pericardial effusion requiring intervention was independently associated with mortality (OR 5.66, 95% CI 4.24-7.56), prolonged length of stay (OR 1.36, 95% CI 1.07-1.73), and increased cost of hospitalization (OR 2.49, 95% CI 1.92-3.21) after leadless pacemaker implantation. CONCLUSION: In a large, contemporary, real-world cohort of leadless pacemaker implantations in the United States, the prevalence of pericardial effusion requiring intervention was 1.1%. Certain important patient-level characteristics predicted development of a significant pericardial effusion, and such effusions were associated with adverse outcomes after leadless pacemaker implantations.

Published
2024

17. Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning

Author

Munir, Muhammad Akhtar, Khan, Muhammad Haris, Sarfraz, M. Saquib, and Ali, Mohsen

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Deep learning based object detectors struggle generalizing to a new target domain bearing significant variations in object and background. Most current methods align domains by using image or instance-level adversarial feature alignment. This often suffers due to unwanted background and lacks class-specific alignment. A straightforward approach to promote class-level alignment is to use high confidence predictions on unlabeled domain as pseudo-labels. These predictions are often noisy since model is poorly calibrated under domain shift. In this paper, we propose to leverage model's predictive uncertainty to strike the right balance between adversarial feature alignment and class-level alignment. We develop a technique to quantify predictive uncertainty on class assignments and bounding-box predictions. Model predictions with low uncertainty are used to generate pseudo-labels for self-training, whereas the ones with higher uncertainty are used to generate tiles for adversarial feature alignment. This synergy between tiling around uncertain object regions and generating pseudo-labels from highly certain object regions allows capturing both image and instance-level context during the model adaptation. We report thorough ablation study to reveal the impact of different components in our approach. Results on five diverse and challenging adaptation scenarios show that our approach outperforms existing state-of-the-art methods with noticeable margins., Comment: Accepted for publication in IEEE Transactions on Pattern Analysis and Machine Intelligence (Volume: 45, Issue: 12, December 2023); Extended version of our conference paper, arXiv link: arXiv:2110.00249

Published
2023

18. Cal-DETR: Calibrated Detection Transformer

Author

Munir, Muhammad Akhtar, Khan, Salman, Khan, Muhammad Haris, Ali, Mohsen, and Khan, Fahad Shahbaz

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Albeit revealing impressive predictive performance for several computer vision tasks, deep neural networks (DNNs) are prone to making overconfident predictions. This limits the adoption and wider utilization of DNNs in many safety-critical applications. There have been recent efforts toward calibrating DNNs, however, almost all of them focus on the classification task. Surprisingly, very little attention has been devoted to calibrating modern DNN-based object detectors, especially detection transformers, which have recently demonstrated promising detection performance and are influential in many decision-making systems. In this work, we address the problem by proposing a mechanism for calibrated detection transformers (Cal-DETR), particularly for Deformable-DETR, UP-DETR and DINO. We pursue the train-time calibration route and make the following contributions. First, we propose a simple yet effective approach for quantifying uncertainty in transformer-based object detectors. Second, we develop an uncertainty-guided logit modulation mechanism that leverages the uncertainty to modulate the class logits. Third, we develop a logit mixing approach that acts as a regularizer with detection-specific losses and is also complementary to the uncertainty-guided logit modulation technique to further improve the calibration performance. Lastly, we conduct extensive experiments across three in-domain and four out-domain scenarios. Results corroborate the effectiveness of Cal-DETR against the competing train-time methods in calibrating both in-domain and out-domain detections while maintaining or even improving the detection performance. Our codebase and pre-trained models can be accessed at \url{https://github.com/akhtarvision/cal-detr}., Comment: Accepted at NeurIPS 2023

Published
2023

26. Development of A Simple 1-T Dew-Point Generator

Author

Tistomo, Arfan Sindhu, Achmadi, Aditya, Fajria, Melati Azizka, Pakpahan, Sri Ningsih Yazana, Arlan, Anastasya, Mareta Dwi, Dewantara, Kelvin Sapta, Azzumar, Muhammad, Prihhapso, Yonan, Ginanjar, Gigin, Viridi, Sparisoma, Munir, Muhammad Miftahul, and Suprijadi

Published
2024
Full Text
View/download PDF

33. Safety and efficacy of colchicine for the prevention of post-operative atrial fibrillation in patients undergoing cardiac surgery: a meta-analysis of randomized controlled trials

Author

Agarwal, Siddharth, Beard, Christopher W, Khosla, Jagjit, Clifton, Shari, Anwaar, Muhammad Faraz, Ghani, Asad, Farhat, Kassem, Pyrpyris, Nikolaos, Momani, Joud, Munir, Muhammad Bilal, DeSimone, Christopher V, Deshmukh, Abhishek, Stavrakis, Stavros, Jackman, Warren M, Po, Sunny, and Asad, Zain Ul Abideen

Subjects
Clinical Research, Cardiovascular, Heart Disease, Clinical Trials and Supportive Activities, Prevention, Patient Safety, Evaluation of treatments and therapeutic interventions, 6.1 Pharmaceuticals, Humans, Colchicine, Atrial Fibrillation, Randomized Controlled Trials as Topic, Cardiac Surgical Procedures, Incidence, Atrial fibrillation, Meta-analysis, Outcomes, Clinical Sciences, Cardiovascular System & Hematology
Abstract

Background and aimsColchicine is an anti-inflammatory drug that may prevent post-operative atrial fibrillation (POAF). The effect of this drug has been inconsistently shown in previous clinical trials. We aimed to compare the efficacy and safety of colchicine vs. placebo to prevent POAF in patients undergoing cardiac surgery.Methods and resultsA systematic search of EMBASE, MEDLINE, SCOPUS, ClinicalTrials.gov, and the Cochrane Library for randomized controlled trials (RCTs) was conducted from inception till April 2023. The primary outcome was the incidence of POAF after any cardiac surgery. The secondary outcome was the rate of drug discontinuation due to adverse events and adverse gastrointestinal events. Risk ratios (RR) were reported using the Mantel Haenszel method. A total of eight RCTs comprising 1885 patients were included. There was a statistically significant lower risk of developing POAF with colchicine vs. placebo (RR: 0.70; 95% CI: 0.59-0.82; P < 0.01, I2 = 0%), and this effect persisted across different subgroups. There was a significantly higher risk of adverse gastrointestinal events (RR: 2.20; 95% CI: 1.38-3.51; P < 0.01, I2 = 55%) with no difference in the risk of drug discontinuation in patients receiving colchicine vs. placebo (RR: 1.33; 95% CI: 0.93-1.89; P = 0.11, I2 = 0%).ConclusionThis meta-analysis of eight RCTs shows that colchicine is effective at preventing POAF, with a significantly higher risk of adverse gastrointestinal events but no difference in the rate of drug discontinuation. Future studies are required to define the optimal duration and dose of colchicine for the prevention of POAF.

Published
2023

34. Intracranial bleeding and associated outcomes in atrial fibrillation patients undergoing percutaneous left atrial appendage occlusion: Insights from National Inpatient Sample 2016-2020.

Author

Khan, Muhammad, Shatla, Islam, Darden, Douglas, Neely, Joseph, Mir, Tanveer, Abideen Asad, Zain, Agarwal, Siddharth, Raina, Sameer, Balla, Sudarshan, Singh, Gagan, Srivatsa, Uma, and Munir, Muhammad

Subjects
Intracranial bleeding, Left atrial appendage occlusion, Mortality, Outcomes
Abstract

BACKGROUND: Percutaneous left atrial appendage occlusion (LAAO) has proved to be a safer alternative for long-term anticoagulation; however, patients with a history of intracranial bleeding were excluded from large randomized clinical trials. OBJECTIVE: The purpose of this study was to determine outcomes in atrial fibrillation (AF) patients with a history of intracranial bleeding undergoing percutaneous LAAO. METHODS: National Inpatient Sample and International Classification of Diseases, Tenth Revision, codes were used to identify patients with AF who underwent LAAO during the years 2016-2020. Patients were stratified based on a history of intracranial bleeding vs not. The outcomes assessed in our study included complications, in-hospital mortality, and resource utilization. RESULT: A total of 89,300 LAAO device implantations were studied. Approximately 565 implantations (0.6%) occurred in patients with a history of intracranial bleed. History of intracranial bleeding was associated with a higher prevalence of overall complications and in-patient mortality in crude analysis. In the multivariate model adjusted for potential confounders, intracranial bleeding was found to be independently associated with in-patient mortality (adjusted odds ratio [aOR] 4.27; 95% confidence interval [CI] 1.68-10.82); overall complications (aOR 1.74; 95% CI 1.36-2.24); prolonged length of stay (aOR 2.38; 95% CI 1.95-2.92); and increased cost of hospitalization (aOR 1.28; 95% CI 1.08-1.52) after percutaneous LAAO device implantation. CONCLUSION: A history of intracranial bleeding was associated with adverse outcomes after percutaneous LAAO. These data, if proven in a large randomized study, can have important clinical consequences in terms of patient selection for LAAO devices.

Published
2023

37. Zoonotic and Zooanthroponotic Potential of Monkeypox

Author

Atasoy, Mustafa O., Naggar, Rania F. El, Rohaim, Mohammed A., Munir, Muhammad, Crusio, Wim E., Series Editor, Dong, Haidong, Series Editor, Radeke, Heinfried H., Series Editor, Rezaei, Nima, Series Editor, Steinlein, Ortrud, Series Editor, and Xiao, Junjie, Series Editor

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results