Search

Your search keyword '"Chinitz, Larry"' showing total 711 results
Start Over Author "Chinitz, Larry"
711 results on '"Chinitz, Larry"'

19. Adversarial Examples for Electrocardiograms

Author

Han, Xintian, Hu, Yuxuan, Foschini, Luca, Chinitz, Larry, Jankelson, Lior, and Ranganath, Rajesh

Subjects
Electrical Engineering and Systems Science - Signal Processing, Computer Science - Cryptography and Security, Computer Science - Machine Learning, Statistics - Machine Learning
Abstract

In recent years, the electrocardiogram (ECG) has seen a large diffusion in both medical and commercial applications, fueled by the rise of single-lead versions. Single-lead ECG can be embedded in medical devices and wearable products such as the injectable Medtronic Linq monitor, the iRhythm Ziopatch wearable monitor, and the Apple Watch Series 4. Recently, deep neural networks have been used to automatically analyze ECG tracings, outperforming even physicians specialized in cardiac electrophysiology in detecting certain rhythm irregularities. However, deep learning classifiers have been shown to be brittle to adversarial examples, which are examples created to look incontrovertibly belonging to a certain class to a human eye but contain subtle features that fool the classifier into misclassifying them into the wrong class. Very recently, adversarial examples have also been created for medical-related tasks. Yet, traditional attack methods to create adversarial examples, such as projected gradient descent (PGD) do not extend directly to ECG signals, as they generate examples that introduce square wave artifacts that are not physiologically plausible. Here, we developed a method to construct smoothed adversarial examples for single-lead ECG. First, we implemented a neural network model achieving state-of-the-art performance on the data from the 2017 PhysioNet/Computing-in-Cardiology Challenge for arrhythmia detection from single lead ECG classification. For this model, we utilized a new technique to generate smoothed examples to produce signals that are 1) indistinguishable to cardiologists from the original examples and 2) incorrectly classified by the neural network. Finally, we show that adversarial examples are not unique and provide a general technique to collate and perturb known adversarial examples to create new ones.

Published
2019

26. Genome-wide association analyses identify new Brugada syndrome risk loci and highlight a new mechanism of sodium channel regulation in disease susceptibility

Author

Barc, Julien, Tadros, Rafik, Glinge, Charlotte, Chiang, David Y., Jouni, Mariam, Simonet, Floriane, Jurgens, Sean J., Baudic, Manon, Nicastro, Michele, Potet, Franck, Offerhaus, Joost A., Walsh, Roddy, Choi, Seung Hoan, Verkerk, Arie O., Mizusawa, Yuka, Anys, Soraya, Minois, Damien, Arnaud, Marine, Duchateau, Josselin, Wijeyeratne, Yanushi D., Muir, Alison, Papadakis, Michael, Castelletti, Silvia, Torchio, Margherita, Ortuño, Cristina Gil, Lacunza, Javier, Giachino, Daniela F., Cerrato, Natascia, Martins, Raphaël P., Campuzano, Oscar, Van Dooren, Sonia, Thollet, Aurélie, Kyndt, Florence, Mazzanti, Andrea, Clémenty, Nicolas, Bisson, Arnaud, Corveleyn, Anniek, Stallmeyer, Birgit, Dittmann, Sven, Saenen, Johan, Noël, Antoine, Honarbakhsh, Shohreh, Rudic, Boris, Marzak, Halim, Rowe, Matthew K., Federspiel, Claire, Le Page, Sophie, Placide, Leslie, Milhem, Antoine, Barajas-Martinez, Hector, Beckmann, Britt-Maria, Krapels, Ingrid P., Steinfurt, Johannes, Winkel, Bo Gregers, Jabbari, Reza, Shoemaker, Moore B., Boukens, Bas J., Škorić-Milosavljević, Doris, Bikker, Hennie, Manevy, Federico C., Lichtner, Peter, Ribasés, Marta, Meitinger, Thomas, Müller-Nurasyid, Martina, Veldink, Jan H., van den Berg, Leonard H., Van Damme, Philip, Cusi, Daniele, Lanzani, Chiara, Rigade, Sidwell, Charpentier, Eric, Baron, Estelle, Bonnaud, Stéphanie, Lecointe, Simon, Donnart, Audrey, Le Marec, Hervé, Chatel, Stéphanie, Karakachoff, Matilde, Bézieau, Stéphane, London, Barry, Tfelt-Hansen, Jacob, Roden, Dan, Odening, Katja E., Cerrone, Marina, Chinitz, Larry A., Volders, Paul G., van de Berg, Maarten P., Laurent, Gabriel, Faivre, Laurence, Antzelevitch, Charles, Kääb, Stefan, Arnaout, Alain Al, Dupuis, Jean-Marc, Pasquie, Jean-Luc, Billon, Olivier, Roberts, Jason D., Jesel, Laurence, Borggrefe, Martin, Lambiase, Pier D., Mansourati, Jacques, Loeys, Bart, Leenhardt, Antoine, Guicheney, Pascale, Maury, Philippe, Schulze-Bahr, Eric, Robyns, Tomas, Breckpot, Jeroen, Babuty, Dominique, Priori, Silvia G., Napolitano, Carlo, de Asmundis, Carlo, Brugada, Pedro, Brugada, Ramon, Arbelo, Elena, Brugada, Josep, Mabo, Philippe, Behar, Nathalie, Giustetto, Carla, Molina, Maria Sabater, Gimeno, Juan R., Hasdemir, Can, Schwartz, Peter J., Crotti, Lia, McKeown, Pascal P., Sharma, Sanjay, Behr, Elijah R., Haissaguerre, Michel, Sacher, Frédéric, Rooryck, Caroline, Tan, Hanno L., Remme, Carol A., Postema, Pieter G., Delmar, Mario, Ellinor, Patrick T., Lubitz, Steven A., Gourraud, Jean-Baptiste, Tanck, Michael W., George, Jr., Alfred L., MacRae, Calum A., Burridge, Paul W., Dina, Christian, Probst, Vincent, Wilde, Arthur A., Schott, Jean-Jacques, Redon, Richard, and Bezzina, Connie R.

Published
2022
Full Text
View/download PDF

28. Performance of a Protein Language Model for Variant Annotation in Cardiac Disease

Author

Hochstadt, Aviram, primary, Barbhaiya, Chirag, additional, Aizer, Anthony, additional, Bernstein, Scott, additional, Cerrone, Marina, additional, Garber, Leonid, additional, Holmes, Douglas, additional, Knotts, Robert J, additional, Kushnir, Alex, additional, Martin, Jacob, additional, Park, David, additional, Spinelli, Michael, additional, Yang, Felix, additional, Chinitz, Larry A, additional, and Jankelson, Lior, additional

Published
2024
Full Text
View/download PDF

37. Predictors of atrial mechanical sensing and atrioventricular synchrony with a leadless ventricular pacemaker: Results from the MARVEL 2 Study

Author

Garweg, Christophe, Khelae, Surinder Kaur, Steinwender, Clemens, Chan, Joseph Yat Sun, Ritter, Philippe, Johansen, Jens Brock, Sagi, Venkata, Epstein, Laurence M., Piccini, Jonathan P., Pascual, Mario, Mont, Lluis, Willems, Rik, Sheldon, Todd, Splett, Vincent, Stromberg, Kurt, Wood, Nicole, and Chinitz, Larry

Published
2020
Full Text
View/download PDF

45. Deep learning models for electrocardiograms are susceptible to adversarial attack

Author

Han, Xintian, Hu, Yuxuan, Foschini, Luca, Chinitz, Larry, Jankelson, Lior, and Ranganath, Rajesh

Subjects
Medical research -- Methods, Medicine, Experimental -- Methods, Electrocardiography -- Research, Electrocardiogram -- Research, Biological sciences, Health
Abstract

Electrocardiogram (ECG) acquisition is increasingly widespread in medical and commercial devices, necessitating the development of automated interpretation strategies. Recently, deep neural networks have been used to automatically analyze ECG tracings and outperform physicians in detecting certain rhythm irregularities.sup.1. However, deep learning classifiers are susceptible to adversarial examples, which are created from raw data to fool the classifier such that it assigns the example to the wrong class, but which are undetectable to the human eye.sup.2,3. Adversarial examples have also been created for medical-related tasks.sup.4,5. However, traditional attack methods to create adversarial examples do not extend directly to ECG signals, as such methods introduce square-wave artefacts that are not physiologically plausible. Here we develop a method to construct smoothed adversarial examples for ECG tracings that are invisible to human expert evaluation and show that a deep learning model for arrhythmia detection from single-lead ECG.sup.6 is vulnerable to this type of attack. Moreover, we provide a general technique for collating and perturbing known adversarial examples to create multiple new ones. The susceptibility of deep learning ECG algorithms to adversarial misclassification implies that care should be taken when evaluating these models on ECGs that may have been altered, particularly when incentives for causing misclassification exist. The development of an algorithm that can imperceptibly manipulate electrocardiographic data to fool a deep learning model for diagnosing cardiac arrhythmia highlights the potential vulnerability of artificial intelligence-enabled diagnosis to adversarial attacks., Author(s): Xintian Han [sup.1] , Yuxuan Hu [sup.2] , Luca Foschini [sup.3] , Larry Chinitz [sup.2] , Lior Jankelson [sup.2] , Rajesh Ranganath [sup.1] [sup.4] [sup.5] Author Affiliations: (1) Center [...]

Published
2020
Full Text
View/download PDF

48. Radiofrequency ablation of drug refractory ventricular tachycardia related to cocaine use: a feasibility, safety, and efficacy study.

Author

Lakkireddy, Dhanunjaya, Kanmanthareddy, Arun, Biria, Mazda, Madhu Reddy, Yeruva, Pillarisetti, Jayasree, Mahapatra, Srijoy, Berenbom, Loren, Chinitz, Larry, Atkins, Donita, Bommana, Sudharani, Tung, Roderick, DI Biase, Luigi, Shivkumar, Kalyanam, and Natale, Andrea

Subjects
cocaine, focal mechanism, myocardial infarction, reentry, ventricular tachycardia, Adolescent, Adult, Aged, Anti-Arrhythmia Agents, Catheter Ablation, Cocaine-Related Disorders, Drug Resistance, Electrophysiologic Techniques, Cardiac, Feasibility Studies, Female, Hemodynamics, Humans, Male, Middle Aged, Recurrence, Registries, Retrospective Studies, Risk Factors, Stroke Volume, Tachycardia, Ventricular, Time Factors, Treatment Outcome, United States, Young Adult
Abstract

BACKGROUND: Cocaine use is a known but rare cause of cardiac arrhythmias. Ventricular arrhythmias related to cocaine may not respond to antiarrhythmic drugs and may need treatment with radiofrequency ablation. OBJECTIVES: We describe the clinical and electrophysiological characteristics of cocaine-related ventricular tachycardia (VT) from a multicenter registry. METHODS: Subjects presenting with VT related to cocaine use and being considered for radiofrequency ablation have been included in the study. Patients who were refractory to maximal medical therapy underwent radiofrequency ablation of the VT. Clinical, procedural variables, efficacy, and safety outcomes were assessed. RESULTS: A total of 14 subjects met study criteria (age 44 ± 13, range 18- to 68-year-old with 79% male, 71% Caucasian). MRI showed evidence of scar only in 43% of patients (6/14). The mechanism of VT was focal in 50% (n = 7) and scar related reentry in 50% (n = 7) based on 3D mapping. The mean VT cycle length was 429 ± 96 milliseconds. The site of origin was epicardial in 16% (3/18) of VTs. Most clinical VTs were hemodynamically stable (75%). Mean ejection fraction at the time of admission was 44 ± 14%. Duration of procedure was 289 ± 50 minutes. One subject developed pericardial tamponade requiring drainage. At 18 ± 11 months follow-up, freedom from arrhythmia was seen in 86% (1 case lost to follow-up and 2 died). CONCLUSION: Radiofrequency ablation is not only feasible but also safe and effective in patients who have drug refractory VT related to chronic cocaine use.

Published
2014

49. Percutaneous left ventricular assist devices in ventricular tachycardia ablation: multicenter experience.

Author

Reddy, Yeruva, Chinitz, Larry, Mansour, Moussa, Bunch, T, Mahapatra, Srijoy, Swarup, Vijay, Di Biase, Luigi, Bommana, Sudharani, Atkins, Donita, Tung, Roderick, Burkhardt, J, Ruskin, Jeremy, Natale, Andrea, Lakkireddy, Dhanunjaya, and Shivkumar, Kalyanam

Subjects
catheter ablation, intra-aortic balloon pumping, tachycardia, ventricular, Aged, Cardiac Catheterization, Catheter Ablation, Echocardiography, Electrocardiography, Female, Follow-Up Studies, Hospital Mortality, Humans, Intra-Aortic Balloon Pumping, Male, Middle Aged, Prospective Studies, Tachycardia, Ventricular, Treatment Outcome, United States, Ventricular Function, Left
Abstract

BACKGROUND: Data on relative safety, efficacy, and role of different percutaneous left ventricular assist devices for hemodynamic support during the ventricular tachycardia (VT) ablation procedure are limited. METHODS AND RESULTS: We performed a multicenter, observational study from a prospective registry including all consecutive patients (N=66) undergoing VT ablation with a percutaneous left ventricular assist devices in 6 centers in the United States. Patients with intra-aortic balloon pump (IABP group; N=22) were compared with patients with either an Impella or a TandemHeart device (non-IABP group; N=44). There were no significant differences in the baseline characteristics between both the groups. In non-IABP group (1) more patients could undergo entrainment/activation mapping (82% versus 59%; P=0.046), (2) more number of unstable VTs could be mapped and ablated per patient (1.05±0.78 versus 0.32±0.48; P

Published
2014

50. Accelerometer-based atrioventricular synchronous pacing with a ventricular leadless pacemaker: Results from the Micra atrioventricular feasibility studies

Author

Chinitz, Larry, Ritter, Philippe, Khelae, Surinder Kaur, Iacopino, Saverio, Garweg, Christophe, Grazia-Bongiorni, Maria, Neuzil, Petr, Johansen, Jens Brock, Mont, Lluis, Gonzalez, Efrain, Sagi, Venkata, Duray, Gabor Z., Clementy, Nicolas, Sheldon, Todd, Splett, Vincent, Stromberg, Kurt, Wood, Nicole, and Steinwender, Clemens

Published
2018
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results