The effect of intrinsic dataset properties on generalization: Unraveling learning differences between natural and medical images
N Konz, MA Mazurowski - arXiv preprint arXiv:2401.08865, 2024 - arxiv.org
This paper investigates discrepancies in how neural networks learn from different imaging
domains, which are commonly overlooked when adopting computer vision techniques from …
domains, which are commonly overlooked when adopting computer vision techniques from …
[HTML][HTML] Neural geometrodynamics, complexity, and plasticity: a psychedelics perspective
We explore the intersection of neural dynamics and the effects of psychedelics in light of
distinct timescales in a framework integrating concepts from dynamics, complexity, and …
distinct timescales in a framework integrating concepts from dynamics, complexity, and …
Attentional decoder networks for chest X-ray image recognition on high-resolution features
Background and objective: This paper introduces an encoder–decoder-based attentional
decoder network to recognize small-size lesions in chest X-ray images. In the encoder-only …
decoder network to recognize small-size lesions in chest X-ray images. In the encoder-only …
Reverse engineering breast mris: Predicting acquisition parameters directly from images
N Konz, MA Mazurowski - Medical Imaging with Deep …, 2024 - proceedings.mlr.press
The image acquisition parameters (IAPs) used to create MRI scans are central to defining
the appearance of the images. Deep learning models trained on data acquired using certain …
the appearance of the images. Deep learning models trained on data acquired using certain …
MiSuRe is all you need to explain your image segmentation
The last decade of computer vision has been dominated by Deep Learning architectures,
thanks to their unparalleled success. Their performance, however, often comes at the cost of …
thanks to their unparalleled success. Their performance, however, often comes at the cost of …
Density estimation via binless multidimensional integration
We introduce the Binless Multidimensional Thermodynamic Integration (BMTI) method for
nonparametric, robust, and data-efficient density estimation. BMTI estimates the logarithm of …
nonparametric, robust, and data-efficient density estimation. BMTI estimates the logarithm of …
Automated Whole-Body Tumor Segmentation and Prognosis of Cancer on PET/CT
KH Leung - Proceedings of the SC'23 Workshops of The …, 2023 - dl.acm.org
Automatic characterization of malignant disease is an important clinical need to facilitate
early detection and treatment of cancer. A deep semi-supervised transfer learning approach …
early detection and treatment of cancer. A deep semi-supervised transfer learning approach …
A Data-Driven Solution for The Cold Start Problem in Biomedical Image Classification
S Kazeminia, M Březík, SS Boushehri… - 2024 IEEE International …, 2024 - ieeexplore.ieee.org
The demand for large quantities of high-quality annotated images poses a significant
bottleneck for developing effective deep learning-based classifiers in the biomedical …
bottleneck for developing effective deep learning-based classifiers in the biomedical …