A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
MobileNet CNN achieves high accuracy in skin disease classification on Android.
The paper investigates skin tone estimation and its impact on dermatology model performance.
In this paper, we demonstrate the potential of applying Variational Autoencoder (VAE) [10] for anomaly detection in skin disease images. VAE is a class of deep generative models which is trained by maximizing the evidence lower bound of data distribution [10]. When trained on only normal data, the resulting model is ab…
Deep learning skin lesion classifier explained using CAVs.
This paper reports the method and evaluation results of MedAusbild team for ISIC challenge task. Since early 2017, our team has worked on melanoma classification [1][6], and has employed deep learning since beginning of 2018 [7]. Deep learning helps researchers absolutely to treat and detect diseases by analyzing medic…
The diagnosis of the Erythemato-squamous disease (ESD) is accepted as a difficult problem in dermatology. ESD is a form of skin disease. It generally causes redness of the skin and also may cause loss of skin. They are generally due to genetic or environmental factors. ESD comprises six classes of skin conditions namel…
CNN automates vitiligo lesion segmentation quickly and accurately.
In this paper, the effectiveness and capability of convolutional neural networks have been studied in the classification of 8 skin diseases. Different pre-trained state-of-the-art architectures (DenseNet 201, ResNet 152, Inception v3, InceptionResNet v2) were used and applied on 10135 dermoscopy skin images in total (H…
This paper introduces a deep-learning based efficient classifier for common dermatological conditions, aimed at people without easy access to skin specialists. We report approximately 80% accuracy, in a situation where primary care doctors have attained 57% success rate, according to recent literature. The rationale of…
Improves AI-prior reliability for Bayesian inference.
Imaging fluorescent disease biomarkers in tissues and skin is a non-invasive method to screen for health conditions. We report an automated process that combines intraoral fluorescent porphyrin biomarker imaging, clinical examinations and machine learning for correlation of systemic health conditions with periodontal d…
A reliable human skin detection method that is adaptable to different human skin colours and illu- mination conditions is essential for better human skin segmentation. Even though different human skin colour detection solutions have been successfully applied, they are prone to false skin detection and are not able to c…
Formula found for skinning map contraction in hyperbolic geometry.
This paper introduces a novel low-cost device prototype for the automatic diagnosis of diseases, utilizing inputted symptoms and personal background. The engineering goal is to solve the problem of limited healthcare access with a single device. Diagnosing diseases automatically is an immense challenge, owing to their …
A statistical toolbox for analyzing model performance in medical imaging.
Certain classes of 3-manifolds, following Thurston, give rise to a 'skinning map', a self-map of the Teichmüller space of the boundary. This paper examines the skinning map of a 3-manifold M, a genus-2 handlebody with two rank-1 cusps. We exploit an orientation-reversing isometry of M to conclude that the skinning map …
Let M be a hyperbolic 3-manifold with nonempty totally geodesic boundary. We prove that there are upper and lower bounds on the diameter of the skinning map of M that depend only on the volume of the hyperbolic structure with totally geodesic boundary, answering a question of Y. Minsky. This is proven via a filling the…
Deep learning improves skin cancer detection.
System detects and classifies skin cancer using CNN.
New method calibrates neural network uncertainty for medical images.
Study reveals AI skin cancer classifiers underperform for darker skin phototypes, advocating for fairness auditing.
This paper proposes an innovative method for segmentation of skin lesions in dermoscopy images developed by the authors, based on fuzzy classification of pixels and histogram thresholding.
Continuum mechanics theory describes skin's complex anisotropic behavior.
This is the first in a series of papers where we develop new structural elements on singular area minimizing hypersurfaces, the skin structures. They disclose previously unapproachable and largely unexpected geometric and analytic properties of such hypersurfaces.
We prove that a Bers slice is never algebraic, meaning that its Zariski closure in the character variety has strictly larger dimension. A corollary is that skinning maps are never constant. The proof uses grafting and the theory of complex projective structures.
We show that the diameter of the skinning map of an acylindrical hyperbolic 3-manifold M is bounded on thick Teichmueller geodesic rays by a constant depending only on the thickness of the ray and the topological type of the boundary of M.
Automated skin lesion analysis is very crucial in clinical practice, as skin cancer is among the most common human malignancy. Existing approaches with deep learning have achieved remarkable performance on this challenging task, however, heavily relying on large-scale labelled datasets. In this paper, we present a nove…
Skin cancer is one of the most common types of cancer around the world. For this reason, over the past years, different approaches have been proposed to assist detect it. Nonetheless, most of them are based only on dermoscopy images and do not take into account the patient clinical information. In this work, first, we …
New method for sensing non-planar surfaces using ERT.
Machine learning improves diagnostic test accuracy for bovine tuberculosis.
We show that if the totally geodesic boundary of a compact hyperbolic 3-manifold M has a large collar of depth d, then the diameter of the skinning map of M is no more than A exp(-d) for some A depending only on the genus and injectivity radius of the boundary of M.
Uniform diameter bound for reflection group disk patterns.
Recent work shows unequal performance of commercial face classification services in the gender classification task across intersectional groups defined by skin type and gender. Accuracy on dark-skinned females is significantly worse than on any other group. In this paper, we conduct several analyses to try to uncover t…
We give a generalization of Thurston's Bounded Image Theorem for skinning maps, which applies to pared 3-manifolds with incompressible boundary that are not necessarily acylindrical. Along the way we study properties of divergent sequences in the deformation space of such a manifold, establishing the existence of compa…
This is the third in a series of papers on the geometry and analysis of singular area minimizing hypersurfaces. We show how to derive obstruction and structure theories for scalar curvature constraints without imposing dimensional or topological restrictions on the underlying manifold. To this end, we use skin structur…
The paper studies circle packings using renormalization and subdivision rules.
In this work, we investigate whether state-of-the-art object detection systems have equitable predictive performance on pedestrians with different skin tones. This work is motivated by many recent examples of ML and vision systems displaying higher error rates for certain demographic groups than others. We annotate an …
Study uses machine learning to optimize antibiotic therapy for MRSA skin infections.
Model patching closes subgroup performance gaps in skin cancer classification.
This paper explains the method used in the segmentation challenge (Task 1) in the International Skin Imaging Collaboration's (ISIC) Skin Lesion Analysis Towards Melanoma Detection challenge held in 2018. We have trained a U-Net network to perform the segmentation. The key elements for the training were first to adjust …
A new SSL method improves medical image classification using global latent mixing.
Study compares DL models for medical image segmentation, finds synergistic ensemble strategies improve performance.
We derive the Riemannian Positive Mass theorem in arbitrary dimensions, without any topological constraints. The main new tools are skin structures and surgeries on minimal hypersurfaces.
Computer-aided diagnosis systems for classification of different type of skin lesions have been an active field of research in recent decades. It has been shown that introducing lesions and their attributes masks into lesion classification pipeline can greatly improve the performance. In this paper, we propose a framew…
We derive the Space-Time Positive Mass theorem in arbitrary dimensions, without topological constraints. The main new tools are skin structures and surgeries on minimal and marginally outer trapped hypersurfaces.
Deep neural network for cancer classification using autoencoders.
This short report describes our submission to the ISIC 2018 Challenge in Skin Lesion Analysis Towards Melanoma Detection for Task1 and Task 3. This work has been accomplished by a team of researchers at the University of Dayton Signal and Image Processing Lab. Our proposed approach is computationally efficient are comb…
Computer Vision and machine learning methods were previously used to reveal screen presence of genders in TV and movies. In this work, using head pose, gender detection, and skin color estimation techniques, we demonstrate that the gender disparity in TV in a South Asian country such as Bangladesh exhibits unique chara…