ObjectiveDeep learning has become a promising approach for automated support for clinical diagnosis. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In this study, we propose methods of distributing deep learning models as an attractive alternative to sharing patient data.MethodsWe simulate the distribution of deep learning models across 4 institutions using various training heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The training heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in 3 independent image collections (retinal fundus photos, mammography, and ImageNet).ResultsWe find that cyclical weight transfer resulted in a performance that was comparable to that of centrally hosted patient data. We also found that there is an improvement in the performance of cyclical weight transfer heuristic with a high frequency of weight transfer.ConclusionsWe show that distributing deep learning models is an effective alternative to sharing patient data. This finding has implications for any collaborative deep learning study.
Cerebral microvascular occlusion is a common phenomenon throughout life1,2 that could be an underappreciated mechanism of brain pathology. Failure to promptly recanalize microvessels may lead to disruption of brain circuits and significant functional deficits3. Hemodynamic forces and the fibrinolytic system4 are considered the principal mechanisms responsible for recanalization of occluded cerebral capillaries and terminal arterioles. However, using high resolution fixed tissue microscopy and two photon imaging in living mice we found that a large fraction of occluding microemboli failed to be lysed and washed out within 48 hours after internal carotid infusion. Surprisingly, emboli were instead found to translocate outside the vessel lumen within 2-7 days leading to complete re-establishment of blood flow and sparing of the vessel. Recanalization occurred by a previously unknown mechanism of microvascular plasticity involving the rapid envelopment of emboli by endothelial membrane projections which subsequently form a new vessel wall. This was followed by the formation of an endothelial opening through which emboli translocated into the perivascular parenchyma. The rate of embolus extravasation was significantly reduced by pharmacological inhibition of matrix metalloproteinase 2/9 activity. In aged mice, extravasation was markedly delayed, resulting in persistent tissue hypoxia, synaptic damage and cell death. Our study identifies a novel cellular mechanism that may be critical for recanalization of occluded microvessels. Alterations in the efficiency of this protective mechanism may have important implications in microvascular pathology, stroke recovery, and age-related cognitive decline.
PurposeTo develop an automated method of localizing and discerning multiple types of findings in retinal images using a limited set of training data without hard-coded feature extraction as a step toward generalizing these methods to rare disease detection in which a limited number of training data are available.MethodsTwo ophthalmologists verified 243 retinal images, labeling important subsections of the image to generate 1324 image patches containing either hemorrhages, microaneurysms, exudates, retinal neovascularization, or normal-appearing structures from the Kaggle dataset. These image patches were used to train one standard convolutional neural network to predict the presence of these five classes. A sliding window method was used to generate probability maps across the entire image.ResultsThe method was validated on the eOphta dataset of 148 whole retinal images for microaneurysms and 47 for exudates. A pixel-wise classification of the area under the curve of the receiver operating characteristic of 0.94 and 0.95, as well as a lesion-wise area under the precision recall curve of 0.86 and 0.64, was achieved for microaneurysms and exudates, respectively.ConclusionsRegionally trained convolutional neural networks can generate lesion-specific probability maps able to detect and distinguish between subtle pathologic lesions with only a few hundred training examples per lesion.
Background Currently, physicians are limited in their ability to provide an accurate prognosis for COVID-19 positive patients. Existing scoring systems have been ineffective for identifying patient decompensation. Machine learning (ML) may offer an alternative strategy. A prospectively validated method to predict the need for ventilation in COVID-19 patients is essential to help triage patients, allocate resources, and prevent emergency intubations and their associated risks. Methods In a multicenter clinical trial, we evaluated the performance of a machine learning algorithm for prediction of invasive mechanical ventilation of COVID-19 patients within 24 h of an initial encounter. We enrolled patients with a COVID-19 diagnosis who were admitted to five United States health systems between March 24 and May 4, 2020. Results 197 patients were enrolled in the REspirAtory Decompensation and model for the triage of covid-19 patients: a prospective studY (READY) clinical trial. The algorithm had a higher diagnostic odds ratio (DOR, 12.58) for predicting ventilation than a comparator early warning system, the Modified Early Warning Score (MEWS). The algorithm also achieved significantly higher sensitivity (0.90) than MEWS, which achieved a sensitivity of 0.78, while maintaining a higher specificity (p < 0.05). Conclusions In the first clinical trial of a machine learning algorithm for ventilation needs among COVID-19 patients, the algorithm demonstrated accurate prediction of the need for mechanical ventilation within 24 h. This algorithm may help care teams effectively triage patients and allocate resources. Further, the algorithm is capable of accurately identifying 16% more patients than a widely used scoring system while minimizing false positive results.
Occlusion of the microvasculature by blood clots, atheromatous fragments, or circulating debris is a frequent phenomenon in most human organs. Emboli are cleared from the microvasculature by hemodynamic pressure and the fibrinolytic system. An alternative mechanism of clearance is angiophagy, in which emboli are engulfed by the endothelium and translocate through the microvascular wall. We report that endothelial lamellipodia surround emboli within hours of occlusion, markedly reducing hemodynamic washout and tissue plasminogen activator-mediated fibrinolysis in mice. Over the next few days, emboli are completely engulfed by the endothelium and extravasated into the perivascular space, leading to vessel recanalization and blood flow reestablishment. We find that this mechanism is not limited to the brain, as previously thought, but also occurs in the heart, retina, kidney, and lung. In the lung, emboli cross into the alveolar space where they are degraded by macrophages, whereas in the kidney, they enter the renal tubules, constituting potential routes for permanent removal of circulating debris. Retina photography and angiography in patients with embolic occlusions provide indirect evidence suggesting that angiophagy may also occur in humans. Thus, angiophagy appears to be a ubiquitous mechanism that could be a therapeutic target with broad implications in vascular occlusive disorders. Given its biphasic nature-initially causing embolus retention, and subsequently driving embolus extravasation-it is likely that different therapeutic strategies will be required during these distinct post-occlusion time windows.
Rationale Prediction of patients at risk for mortality can help triage patients and assist in resource allocation. Objectives Develop and evaluate a machine learning-based algorithm which accurately predicts mortality in COVID-19, pneumonia, and mechanically ventilated patients. Methods Retrospective study of 53,001 total ICU patients, including 9166 patients with pneumonia and 25,895 mechanically ventilated patients, performed on the MIMIC dataset. An additional retrospective analysis was performed on a community hospital dataset containing 114 patients positive for SARS-COV-2 by PCR test. The outcome of interest was in-hospital patient mortality. Results When trained and tested on the MIMIC dataset, the XGBoost predictor obtained area under the receiver operating characteristic (AUROC) values of 0.82, 0.81, 0.77, and 0.75 for mortality prediction on mechanically ventilated patients at 12-, 24-, 48-, and 72- hour windows, respectively, and AUROCs of 0.87, 0.78, 0.77, and 0.734 for mortality prediction on pneumonia patients at 12-, 24-, 48-, and 72- hour windows, respectively. The predictor outperformed the qSOFA, MEWS and CURB-65 risk scores at all prediction windows. When tested on the community hospital dataset, the predictor obtained AUROCs of 0.91, 0.90, 0.86, and 0.87 for mortality prediction on COVID-19 patients at 12-, 24-, 48-, and 72- hour windows, respectively, outperforming the qSOFA, MEWS and CURB-65 risk scores at all prediction windows. Conclusions This machine learning-based algorithm is a useful predictive tool for anticipating patient mortality at clinically useful, and is capable of accurate mortality prediction for mechanically ventilated patients as well as those diagnosed with pneumonia and COVID-19.
BCVA measured by non-ophthalmic ED staff with an app was more accurate than with a Snellen chart. Automated apps may provide a means to standardize and improve the efficiency of ED ophthalmologic care.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.