no code implementations • ICLR 2019 • Saeid Asgari Taghanaki, Shekoofeh Azizi, Ghassan Hamarneh
The linear and non-flexible nature of deep convolutional models makes them vulnerable to carefully crafted adversarial perturbations.
1 code implementation • 13 Mar 2024 • Ashish Sinha, Ghassan Hamarneh
We propose a novel approach for representing anatomical trees using INR, while also capturing the distribution of a set of trees via denoising diffusion in the space of INRs.
no code implementations • 7 Mar 2024 • Adriano D'Alessandro, Ali Mahdavi-Amiri, Ghassan Hamarneh
Consequently, we can generate counting data for any type of object and count them in an unsupervised manner.
1 code implementation • 25 Jan 2024 • Kumar Abhishek, Aditi Jain, Ghassan Hamarneh
The remarkable progress of deep learning in dermatological tasks has brought us closer to achieving diagnostic accuracies comparable to those of human experts.
1 code implementation • 2 Oct 2023 • Adriano D'Alessandro, Ali Mahdavi-Amiri, Ghassan Hamarneh
To address this, we use latent diffusion models to create two types of synthetic data: one by removing pedestrians from real images, which generates ranked image pairs with a weak but reliable object quantity signal, and the other by generating synthetic images with a predetermined number of objects, offering a strong but noisy counting signal.
1 code implementation • 6 Sep 2023 • Aliasghar Khani, Saeid Asgari Taghanaki, Aditya Sanghi, Ali Mahdavi Amiri, Ghassan Hamarneh
Then, using the extracted attention maps, the text embeddings of Stable Diffusion are optimized such that, each of them, learn about a single segmented region from the training image.
no code implementations • 26 May 2023 • Ivan R. Nabi, Ben Cardoen, Ismail M. Khater, Guang Gao, Timothy H. Wong, Ghassan Hamarneh
The nanoscale resolution of super-resolution microscopy has now enabled the use of fluorescent based molecular localization tools to study whole cell structural biology.
1 code implementation • 22 May 2023 • Ashish Sinha, Jeremy Kawahara, Arezou Pakzad, Kumar Abhishek, Matthieu Ruthven, Enjie Ghorbel, Anis Kacem, Djamila Aouada, Ghassan Hamarneh
In recent years, deep learning (DL) has shown great potential in the field of dermatological image analysis.
no code implementations • 30 Mar 2023 • Weina Jin, Xiaoxiao Li, Ghassan Hamarneh
Optimizing XAI for plausibility regardless of the model decision correctness also jeopardizes model trustworthiness, because doing so breaks an important assumption in human-human explanation that plausible explanations typically imply correct decisions, and vice versa; and violating this assumption eventually leads to either undertrust or overtrust of AI models.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI) +1
1 code implementation • 10 Feb 2023 • Weina Jin, Jianyu Fan, Diane Gromala, Philippe Pasquier, Ghassan Hamarneh
The EUCA study findings, the identified explanation forms and goals for technical specification, and the EUCA study dataset support the design and evaluation of end-user-centered XAI techniques for accessible, safe, and accountable AI.
1 code implementation • 30 Sep 2022 • Saeid Asgari Taghanaki, Aliasghar Khani, Fereshte Khani, Ali Gholami, Linh Tran, Ali Mahdavi-Amiri, Ghassan Hamarneh
A fundamental challenge of over-parameterized deep learning models is learning meaningful data representations that yield good performance on a downstream task without over-fitting spurious input features.
1 code implementation • 29 Aug 2022 • Arezou Pakzad, Kumar Abhishek, Ghassan Hamarneh
While deep learning based approaches have demonstrated expert-level performance in dermatological diagnosis tasks, they have also been shown to exhibit biases toward certain demographic attributes, particularly skin types (e. g., light versus dark), a fairness concern that must be addressed.
1 code implementation • 22 Aug 2022 • Siyi Du, Ben Hers, Nourhan Bayasi, Ghassan Hamarneh, Rafeef Garbi
Deep learning models have achieved great success in automating skin lesion diagnosis.
no code implementations • 18 Aug 2022 • Weina Jin, Jianyu Fan, Diane Gromala, Philippe Pasquier, Xiaoxiao Li, Ghassan Hamarneh
The boundaries of existing explainable artificial intelligence (XAI) algorithms are confined to problems grounded in technical users' demand for explainability.
1 code implementation • 1 Jun 2022 • Zahra Mirikharaji, Kumar Abhishek, Alceu Bissoto, Catarina Barata, Sandra Avila, Eduardo Valle, M. Emre Celebi, Ghassan Hamarneh
We analyze these works along several dimensions, including input data (datasets, preprocessing, and synthetic data generation), model design (architecture, modules, and losses), and evaluation aspects (data annotation requirements and segmentation performance).
no code implementations • 7 Apr 2022 • Kumar Abhishek, Colin J. Brown, Ghassan Hamarneh
Modern deep learning training procedures rely on model regularization techniques such as data augmentation methods, which generate training samples that increase the diversity of data and richness of label information.
1 code implementation • 12 Mar 2022 • Weina Jin, Xiaoxiao Li, Ghassan Hamarneh
The evaluation and MSFI metric can guide the design and selection of XAI algorithms to meet clinical requirements on multi-modal explanation.
Explainable Artificial Intelligence (XAI) Feature Importance
1 code implementation • 10 Mar 2022 • Mayur Mallya, Ghassan Hamarneh
Furthermore, in the case of brain tumor classification, our method outperforms the model trained on the superior modality while producing comparable results to the model that uses both modalities during inference.
1 code implementation • 16 Feb 2022 • Weina Jin, Xiaoxiao Li, Mostafa Fatehi, Ghassan Hamarneh
Following the guidelines, we conducted a systematic evaluation on a novel problem of multi-modal medical image explanation with two clinical tasks, and proposed new evaluation metrics accordingly.
Computational Efficiency Explainable artificial intelligence +1
no code implementations • CVPR 2022 • Nourhan Bayasi, Ghassan Hamarneh, Rafeef Garbi
Deep learning (DL) models trained to minimize empirical risk on a single domain often fail to generalize when applied to other domains.
no code implementations • 29 Sep 2021 • Adriano C. D'Alessandro, Ali Mahdavi Amiri, Ghassan Hamarneh
Object counting and localization in dense scenes is a challenging class of image analysis problems that typically requires labour intensive annotations to learn to solve.
no code implementations • 11 Jul 2021 • Weina Jin, Xiaoxiao Li, Ghassan Hamarneh
The maps highlight important features for AI model's prediction.
2 code implementations • 10 May 2021 • M. Sadegh Saberian, Kathleen P. Moriarty, Andrea D. Olmstead, Christian Hallgrimson, François Jean, Ivan R. Nabi, Maxwell W. Libbrecht, Ghassan Hamarneh
DEEMD can be explored for use on other emerging viruses and datasets to rapidly identify candidate antiviral treatments in the future}.
1 code implementation • 2 May 2021 • Mengliu Zhao, Jeremy Kawahara, Kumar Abhishek, Sajjad Shamanian, Ghassan Hamarneh
Our lesion tracking algorithm achieves an average matching accuracy of 88% on a set of detected corresponding pairs of prominent lesions of subjects imaged in different poses, and an average longitudinal accuracy of 71% when encompassing additional errors due to lesion detection.
1 code implementation • 4 Feb 2021 • Weina Jin, Jianyu Fan, Diane Gromala, Philippe Pasquier, Ghassan Hamarneh
The ability to explain decisions to end-users is a necessity to deploy AI as critical decision support.
Decision Making Explainable artificial intelligence Human-Computer Interaction
no code implementations • 14 Dec 2020 • Zahra Mirikharaji, Kumar Abhishek, Saeed Izadi, Ghassan Hamarneh
To this end, we propose an ensemble of Bayesian fully convolutional networks (FCNs) for the segmentation task by considering two major factors in the aggregation of multiple ground truth annotations: (1) handling contradictory annotations in the training data originating from inter-annotator disagreements and (2) improving confidence calibration through the fusion of base models' predictions.
1 code implementation • 26 Oct 2020 • Kumar Abhishek, Ghassan Hamarneh
The segmentation of skin lesions is a crucial task in clinical decision support systems for the computer aided diagnosis of skin lesions.
no code implementations • 25 Mar 2020 • Saeed Izadi, Ghassan Hamarneh
The performance of our proposed method is evaluated on confocal microscopy images with real noise Poisson-Gaussian noise.
1 code implementation • 23 Mar 2020 • Kumar Abhishek, Ghassan Hamarneh, Mark S. Drew
The semantic segmentation of skin lesions is an important and common initial task in the computer aided diagnosis of dermoscopic images.
no code implementations • 26 Feb 2020 • Hanene Ben Yedder, Ben Cardoen, Ghassan Hamarneh
Medical imaging is an invaluable resource in medicine as it enables to peer inside the human body and provides scientists and physicians with a wealth of information indispensable for understanding, modelling, diagnosis, and treatment of diseases.
no code implementations • 28 Nov 2019 • Weina Jin, Mostafa Fatehi, Kumar Abhishek, Mayur Mallya, Brian Toyota, Ghassan Hamarneh
We believe that these technical approaches will facilitate the development of a fully-functional AI tool in the clinical care of patients with gliomas.
no code implementations • 16 Nov 2019 • Saeid Asgari Taghanaki, Kumar Abhishek, Ghassan Hamarneh
To test the effectiveness of the proposed idea and compare it with other competing methods, we design several test scenarios, such as classification performance, uncertainty, out-of-distribution, and robustness analyses.
no code implementations • 27 Oct 2019 • Aïcha BenTaieb, Ghassan Hamarneh
We then discuss the challenges facing the validation and integration of such deep learning-based computational systems in clinical workflow and reflect on future opportunities for histopathology derived image measurements and better predictive modeling.
no code implementations • 22 Oct 2019 • Shahab Aslani, Vittorio Murino, Michael Dayan, Roger Tam, Diego Sona, Ghassan Hamarneh
This paper presents a simple and effective generalization method for magnetic resonance imaging (MRI) segmentation when data is collected from multiple MRI scanning sites and as a consequence is affected by (site-)domain shifts.
no code implementations • 16 Oct 2019 • Saeid Asgari Taghanaki, Kumar Abhishek, Joseph Paul Cohen, Julien Cohen-Adad, Ghassan Hamarneh
The semantic image segmentation task consists of classifying each pixel of an image into an instance, where each instance corresponds to a class.
no code implementations • 8 Aug 2019 • Saeed Izadi, Zahra Mirikharaji, Mengliu Zhao, Ghassan Hamarneh
Specifically, by using total variation and piecewise constancy priors along with noise whiteness priors such as auto-correlation and stationary losses, our network learns to decouple an input noisy image into the underlying signal and noise components.
no code implementations • 18 Jun 2019 • Saeed Izadi, Darren Sutton, Ghassan Hamarneh
We compare the efficacy of our method to 11 other existing single image super resolution techniques that compensate for the reduction in image quality caused by the necessity of endomicroscope miniaturization.
no code implementations • 13 Jun 2019 • Kumar Abhishek, Ghassan Hamarneh
Skin lesion segmentation is a vital task in skin cancer diagnosis and further treatment.
no code implementations • 10 Jun 2019 • Zahra Mirikharaji, Yiqi Yan, Ghassan Hamarneh
Deep convolutional neural networks have driven substantial advancements in the automatic understanding of images.
no code implementations • 4 Jun 2019 • Jeremy Kawahara, Ghassan Hamarneh
Skin conditions are a global health concern, ranking the fourth highest cause of nonfatal disease burden when measured as years lost due to disability.
no code implementations • 27 Apr 2019 • Anmol Sharma, Ghassan Hamarneh
The ability to visualize tissue in varied contrasts in the form of MR pulse sequences in a single scan provides valuable insights to physicians, as well as enabling automated systems performing downstream analysis.
no code implementations • 4 Apr 2019 • Saied Asgari Taghanaki, Kumar Abhishek, Ghassan Hamarneh
Although numerous improvements have been made in the field of image segmentation using convolutional neural networks, the majority of these improvements rely on training with larger datasets, model architecture modifications, novel loss functions, and better optimizers.
no code implementations • 28 Mar 2019 • Saeid Asgari Taghanaki, Mohammad Havaei, Tess Berthier, Francis Dutil, Lisa Di Jorio, Ghassan Hamarneh, Yoshua Bengio
The scarcity of richly annotated medical images is limiting supervised deep learning based solutions to medical image analysis tasks, such as localizing discriminatory radiomic disease signatures.
1 code implementation • CVPR 2019 • Saeid Asgari Taghanaki, Kumar Abhishek, Shekoofeh Azizi, Ghassan Hamarneh
The linear and non-flexible nature of deep convolutional models makes them vulnerable to carefully crafted adversarial perturbations.
no code implementations • 9 Jul 2018 • Saeid Asgari Taghanaki, Arkadeep Das, Ghassan Hamarneh
Recently, there have been several successful deep learning approaches for automatically classifying chest X-ray images into different disease categories.
no code implementations • 21 Jun 2018 • Saeed Izadi, Kathleen P. Moriarty, Ghassan Hamarneh
In this work, we demonstrate that software-based techniques can be used to recover lost information due to endomicroscopy hardware miniaturization and reconstruct images of higher resolution.
no code implementations • 21 Jun 2018 • Zahra Mirikharaji, Ghassan Hamarneh
Semantic segmentation is an important preliminary step towards automatic medical image interpretation.
1 code implementation • 8 May 2018 • Saeid Asgari Taghanaki, Yefeng Zheng, S. Kevin Zhou, Bogdan Georgescu, Puneet Sharma, Daguang Xu, Dorin Comaniciu, Ghassan Hamarneh
The output imbalance refers to the imbalance between the false positives and false negatives of the inference model.
no code implementations • 14 Apr 2018 • Saeid Asgari Taghanaki, Aicha Bentaieb, Anmol Sharma, S. Kevin Zhou, Yefeng Zheng, Bogdan Georgescu, Puneet Sharma, Sasa Grbic, Zhoubing Xu, Dorin Comaniciu, Ghassan Hamarneh
Skip connections in deep networks have improved both segmentation and classification performance by facilitating the training of deeper network architectures, and reducing the risks for vanishing gradients.
no code implementations • 14 Mar 2017 • Jeremy Kawahara, Ghassan Hamarneh
We reformulate the task of classifying clinical dermoscopic features within superpixels as a segmentation problem, and propose a fully convolutional neural network to detect clinical dermoscopic features from dermoscopy skin lesion images.
no code implementations • 26 Nov 2016 • Colin J Brown, Ghassan Hamarneh
The purpose of this work is to review the literature on the topic of applying machine learning models to MRI-based connectome data.
no code implementations • 14 Jul 2016 • Shawn Andrews, Ghassan Hamarneh
Specifically, we dynamically determine the number of eigenvectors needed for a desired accuracy based on user input, and derive update equations for the eigenvectors when the edge weights or topology of the image graph are changed.
no code implementations • 5 Jul 2016 • Masoud S. Nosrati, Ghassan Hamarneh
Medical image segmentation, the task of partitioning an image into meaningful parts, is an important step toward automating medical image analysis and is at the crux of a variety of medical imaging applications, such as computer aided diagnosis, therapy planning and delivery, and computer aided interventions.
no code implementations • 24 Sep 2015 • Shawn Andrews, Ghassan Hamarneh
Probabilistic image segmentation is increasingly popular as it captures uncertainty in the results.