A Patient-Centric Dataset of Images and Metadata for Identifying Melanomas Using Clinical Context

Rotemberg, Veronica; Kurtansky, Nicholas; Betz‐Stablein, Brigid; Caffery, Liam J; Chousakos, Emmanouil; Codella, Noel; Combalia, Marc; Dusza, Stephen W.; Guitera, Pascale; Gutman, David; Halpern, Allan C.; Kittler, Harald; Kose, Katsumi; Langer, Steve G.; Lioprys, Konstantinos; Malvehy, Josep; Musthaq, Shenara; Nanda, Jabpani; Reiter, Ofer; Shih, George; Stratigos, Alexander J.; Tschandl, Philipp; Weber, Jochen; Soyer, H. Peter

doi:10.48550/arxiv.2008.07360

Cited by 6 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data. To further demonstrate the advantages of our method, we conduct experiments on the Melanoma dataset Rotemberg et al (2020) from a Kaggle competition. The dataset contains 33,126 labeled medical images, among which 584 images are related to malignant melanoma and labelled as positive samples.…”

Section: Image Classificationmentioning

confidence: 99%

Stochastic Optimization of Areas UnderPrecision-Recall Curves with Provable Convergence

Luo

Zhao

et al. 2021

Preprint

View full text Add to dashboard Cite

Areas under ROC (AUROC) and precision-recall curves (AUPRC) are common metrics for evaluating classification performance for imbalanced problems. Compared with AUROC, AUPRC is a more appropriate metric for highly imbalanced datasets. While direct optimization of AUROC has been studied extensively, optimization of AUPRC has been rarely explored. In this work, we propose a principled technical method to optimize AUPRC for deep learning. Our approach is based on maximizing the averaged precision (AP), which is an unbiased point estimator of AUPRC. We show that the surrogate loss function for AP is highly non-convex and more complicated than that of AUROC. We cast the objective into a sum of dependent compositional functions with inner functions dependent on random variables of the outer level. We propose efficient adaptive and non-adaptive stochastic algorithms with provable convergence guarantee under mild conditions by using recent advances in stochastic compositional optimization. Extensive experimental results on graph and image datasets demonstrate that our proposed method outperforms prior methods on imbalanced problems. To the best of our knowledge, our work represents the first attempt to optimize AUPRC with provable convergence. 1 1. * denotes equal contribution.

show abstract

Section: Image Classificationmentioning

confidence: 99%

Stochastic Optimization of Areas UnderPrecision-Recall Curves with Provable Convergence

Luo

Zhao

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…For chest X-ray and CT-MRI liver segmentation, we report per-patient DICE scores, the default metric used in the literature for this task. For skin lesion segmentation, we report the per-patient JC index, following the ISIC challenge [79]. For face part segmentation, we use mean Intersection over Union (mIoU) over all classes, excluding the background class.…”

Section: Setupmentioning

confidence: 99%

Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization

Li¹,

Yang²,

Kreis³

et al. 2021

Preprint

View full text Add to dashboard Cite

Training deep networks with limited labeled data while achieving a strong generalization ability is key in the quest to reduce human annotation efforts. This is the goal of semi-supervised learning, which exploits more widely available unlabeled data to complement small labeled data sets. In this paper, we propose a novel framework for discriminative pixel-level tasks using a generative model of both images and labels. Concretely, we learn a generative adversarial network that captures the joint image-label distribution and is trained efficiently using a large set of unlabeled images supplemented with only few labeled ones. We build our architecture on top of StyleGAN2 [45], augmented with a label synthesis branch. Image labeling at test time is achieved by first embedding the target image into the joint latent space via an encoder network and testtime optimization, and then generating the label from the inferred embedding. We evaluate our approach in two important domains: medical image segmentation and part-based face segmentation. We demonstrate strong in-domain performance compared to several baselines, and are the first to showcase extreme out-of-domain generalization, such as transferring from CT to MRI in medical imaging, and photographs of real faces to paintings, sculptures, and even cartoons and animal faces. Project

show abstract

“…Wu et al develop a computer vision model, finding it as accurate as experienced radiologists in screening breast cancer [12]. Moreover, there are a number of benchmark datasets that have shown impressive improvements in arXiv:2105.02386v1 [q-bio.QM] 6 May 2021 performance for disease diagnosis, including CheXpert [4], SD-198 [11], and the International Skin Imaging Collaboration (ISIC) dataset [9].…”

Section: A Machine Learning In Health Informaticsmentioning

confidence: 99%

Automated Primary Hyperparathyroidism Screening with Neural Networks

Ziems

Norman³

2021

2021 IEEE Global Communications Conference (GLOBECOM)

View full text Add to dashboard Cite

Primary Hyperparathyroidism(PHPT) is a relatively common disease, affecting about one in every 1,000 adults. However, screening for PHPT can be difficult, meaning it often goes undiagnosed for long periods of time. While looking at specific blood test results independently can help indicate whether a patient has PHPT, often these blood result levels can all be within their respective normal ranges despite the patient having PHPT. Based on the clinic data from the real world, in this work, we propose a novel approach to screening PHPT with neural network (NN) architecture, achieving over 97% accuracy with common blood values as inputs. Further, we propose a second model achieving over 99% accuracy with additional lab test values as inputs. Moreover, compared to traditional PHPT screening methods, our NN models can reduce the false negatives of traditional screening methods by 99%.

show abstract

A Patient-Centric Dataset of Images and Metadata for Identifying Melanomas Using Clinical Context

Cited by 6 publications

References 0 publications

Stochastic Optimization of Areas UnderPrecision-Recall Curves with Provable Convergence

Stochastic Optimization of Areas UnderPrecision-Recall Curves with Provable Convergence

Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization

Automated Primary Hyperparathyroidism Screening with Neural Networks

Contact Info

Product

Resources

About