Filters








8,046 Hits in 9.8 sec

Towards Intrinsic Interactive Reinforcement Learning [article]

Benjamin Poole, Minwoo Lee
2022 arXiv   pre-print
These two ideas have set RL and BCI on a collision course for one another through the integration of BCI into the IRL framework where intrinsic feedback can be utilized to help train an agent.  ...  With the rising interest in human-in-the-loop (HITL) applications, RL algorithms have been adapted to account for human guidance giving rise to the sub-field of interactive reinforcement learning (IRL)  ...  ACKNOWLEDGMENTS The authors would like to especially thank José del R. Millán, Brad Knox, and Peter Stone for taking the time to provide valuable feedback.  ... 
arXiv:2112.01575v2 fatcat:orpakgcvbrdh5ddhpmpbtxneau

Guest Editorial Annotation-Efficient Deep Learning: The Holy Grail of Medical Imaging

Nima Tajbakhsh, Holger Roth, Demetri Terzopoulos, Jianming Liang
2021 IEEE Transactions on Medical Imaging  
Los Angeles, CA 90024 USA JIANMING LIANG , Guest Editor Biomedical Informatics Program Arizona State University Tempe, AZ 85281 USA APPENDIX RELATED WORKS [ing uncertainty within the training pool: Active  ...  Active learning also may be embedded within interactive segmentation to suggest which parts of the image should be segmented next, thereby further accelerating the annotation process.  ... 
doi:10.1109/tmi.2021.3089292 pmid:34795461 pmcid:PMC8594751 fatcat:t7kufjbdyfgazng3gcuyuhawxu

Deep Learning: An Update for Radiologists

Phillip M. Cheng, Emmanuel Montagnon, Rikiya Yamashita, Ian Pan, Alexandre Cadrin-Chênevert, Francisco Perdigón Romero, Gabriel Chartrand, Samuel Kadoury, An Tang
2021 Radiographics  
Familiarity with the key concepts described will help radiologists understand advances of deep learning in medical imaging and facilitate clinical adoption of these techniques.  ...  Unlike traditional machine learning methods that require hand-engineered feature extraction from input images, deep learning methods learn the image features by which to classify data.  ...  -The U-Net is a popular architecture originally developed for segmenting microscopy images (46) but which continues to be widely used both within and outside the medical domain.  ... 
doi:10.1148/rg.2021200210 pmid:34469211 fatcat:rr3rxpmsbbd7th42w6hfiqc3uy

Deep Learning in Image Cytometry: A Review

Anindya Gupta, Philip J. Harrison, Håkan Wieslander, Nicolas Pielawski, Kimmo Kartasalo, Gabriele Partel, Leslie Solorzano, Amit Suveer, Anna H. Klemm, Ola Spjuth, Ida‐Maria Sintorn, Carolina Wählby
2018 Cytometry Part A  
for extracting information from image data.  ...  We do not provide a full manual for applying these methods to your own data, but rather review previously published articles on deep learning in image cytometry, and guide the readers toward further reading  ...  ACKNOWLEDGMENTS This project was financially supported by the Swedish Foundation for Strategic Research (grants BD15-0008 and SB16-0046), the European Research Council (grant ERC-2015-CoG 682810), and  ... 
doi:10.1002/cyto.a.23701 pmid:30565841 pmcid:PMC6590257 fatcat:dszbcsfncrhxnazsxopjkbe3ju

Urban Corridors as Common Pool Resources: the Case of Nova Gorica and Rijeka

Marco Acri, Saša Dobričić, Maja Debevec
2021 Zenodo  
The concerns for climate change and the increase of urban population made cities fundamental in finding solutions for a more sustainable dwelling to increase resilience and quality of life.  ...  Thanks to the co-creation processes that have been initiated, both corridors are expected to become new social common places for the well-being and represent a paradigm for the sustainable regeneration  ...  The sport playgrounds were marked as places for active physical training and received positive evaluation, too.  ... 
doi:10.5281/zenodo.5592101 fatcat:a6ywxtrbp5esjeeap7bqfb4r5i

Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models

Jialin Peng, Ye Wang
2021 IEEE Access  
The labeling costs for medical images are very high, especially in medical image segmentation, which typically requires intensive pixel/voxel-wise labeling.  ...  application of deep learning models in medical image segmentation.  ...  After that, the more complex pixel-level FIGURE 6. The flowchart of the self-training for semi-supervised segmentation and active learning for interactive segmentation on a conceptual level.  ... 
doi:10.1109/access.2021.3062380 fatcat:r5vsec2yfzcy5nk7wusiftyayu

Accuracy, Uncertainty, and Adaptability of Automatic Myocardial ASL Segmentation using Deep CNN [article]

Hung P. Do, Yi Guo, Andrew J. Yoon, Krishna S. Nayak
2019 arXiv   pre-print
PURPOSE: To apply deep CNN to the segmentation task in myocardial arterial spin labeled (ASL) perfusion imaging and to develop methods that measure uncertainty and that adapt the CNN model to a specific  ...  CONCLUSION: We demonstrate the feasibility of deep CNN for automatic segmentation of myocardial ASL, with good accuracy. We also introduce two simple methods for assessing model uncertainty.  ...  ACKNOWLEDGEMENTS This study was supported by NIH Grant R01HL130494-01A1 and the Whittier Foundation # 0003457-00001  ... 
arXiv:1812.03974v4 fatcat:w24ieydm65cnzlcmatc3hecq54

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

Abdullah A. Abdullah, Masoud M. Hassan, Yaseen T. Mustafa
2022 IEEE Access  
In particular, the applications of Bayesian deep learning in healthcare have been discussed such as its use in medical imaging tasks, clinical signal processing, medical natural language processing, and  ...  Unlike classical DL, BDL uses probability distributions for the model parameters, which makes it possible to estimate the whole uncertainties associated with the predicted outputs.  ...  [60] proposed an active learning method based on BDL U-Net architecture for image segmentation.  ... 
doi:10.1109/access.2022.3163384 fatcat:bp3cwmcbazc63izsbz7e2tpxku

CorneaNet: fast segmentation of cornea OCT scans of healthy and keratoconic eyes using deep learning

Valentin Aranha dos Santos, Leopold Schmetterer, Hannes Stegmann, Martin Pfister, Alina Messner, Gerald Schmidinger, Gerhard Garhofer, René M. Werkmeister
2019 Biomedical Optics Express  
In total, 20,160 images were labeled and used for the training in a supervised learning approach. A custom neural network architecture called CorneaNet was designed and trained.  ...  Our results show that deep learning algorithms can be used for OCT image segmentation and could be applied in various clinical settings.  ...  Niklas Pircher of the Department of Ophthalmology of the Medical University of Vienna for his help for the recruitment and measurement of patients with keratoconus.  ... 
doi:10.1364/boe.10.000622 pmid:30800504 pmcid:PMC6377876 fatcat:hyhtt64ic5fd3m6epcvi5smyuq

Deep learning in medical imaging and radiation therapy

Berkman Sahiner, Aria Pezeshk, Lubomir M. Hadjiiski, Xiaosong Wang, Karen Drukker, Kenny H. Cha, Ronald M. Summers, Maryellen L. Giger
2018 Medical Physics (Lancaster)  
The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and strategies  ...  for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.  ...  Mosinska et al. 344 tailored the uncertainty sampling-based active learning approach for the delineation of complex linear structures problem, which significantly reduced the size (up to 80%) of training  ... 
doi:10.1002/mp.13264 pmid:30367497 fatcat:bottst5mvrbkfedbuocbrstcnm

SegChaNet: A Novel Model for Lung Cancer Segmentation in CT Scans

Mehmet Akif Cifci, Yaodong Gu
2022 Applied Bionics and Biomechanics  
Due to the low contrast of the lung tumor in computed tomography (CT) images, segmentation of the tumor in CT images is challenging.  ...  Accurate lung tumor identification is crucial for radiation treatment planning.  ...  Using medical imaging may be possible to develop more accurate models for multiorgan segmentation.  ... 
doi:10.1155/2022/1139587 pmid:35607427 pmcid:PMC9124150 fatcat:v3yw7ydz2zbpzgkdt635pitjsq

Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future [article]

David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, Lars Petersson
2021 arXiv   pre-print
We also outline the limitations of existing techniques and discuss potential directions for future research.  ...  With the advances of data-driven machine learning research, a wide variety of prediction problems have been tackled.  ...  L 2 -GCN is also introduced where an RNN controller learns a stopping criteria for each layer trained within the L-GCN.  ... 
arXiv:2105.13137v1 fatcat:gm7d2ziagba7bj3g34u4t3k43y

Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future

David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, Lars Petersson
2021 Sensors  
We also outline the limitations of existing techniques and discuss potential directions for future research.  ...  With the advances of data-driven machine learning research, a wide variety of prediction problems have been tackled.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s21144758 fatcat:jytyt4u2pjgvhnhcto3vcvd3a4

UACANet: Uncertainty Augmented Context Attention for Polyp Segmentation [article]

Taehun Kim, Hyemin Lee, Daijin Kim
2021 arXiv   pre-print
We propose Uncertainty Augmented Context Attention network (UACANet) for polyp segmentation which consider a uncertain area of the saliency map.  ...  We conduct experiments on five popular polyp segmentation benchmarks, Kvasir, CVC-ClinicDB, ETIS, CVC-ColonDB and CVC-300, and achieve state-of-the-art performance.  ...  Medical image segmentation is widely used technique such as classifying each organ in the given tomography images like pancreas segmentation [20] , detecting cells from the microscopy images [22] , or  ... 
arXiv:2107.02368v2 fatcat:cczi3jvt45bbpj3dzaaggyjr2m

Class-Balanced Active Learning for Image Classification [article]

Javad Zolfaghari Bengar, Joost van de Weijer, Laura Lopez Fuentes, Bogdan Raducanu
2021 arXiv   pre-print
Active learning aims to reduce the labeling effort that is required to train algorithms by learning an acquisition function selecting the most relevant data for which a label should be requested from a  ...  We argue that this further complicates the active learning process, since the imbalanced data pool can result in suboptimal classifiers.  ...  Acknowledgements We ackowledge the support of the project PID2019-104174GB-I00 (MINECO, Spain), the CERCA Programme of Generalitat de Catalunya, the EU project CybSpeed MSCA-RISE-2017-777720 and CYTED  ... 
arXiv:2110.04543v1 fatcat:kv2bimznpjc2dov6yhr7ba7spq
« Previous Showing results 1 — 15 out of 8,046 results