ABSTRACT
We seek to transform how new and emergent variants of pandemic-causing viruses, specifically SARS-CoV-2, are identified and classified. By adapting large language models (LLMs) for genomic data, we build genome-scale language models (GenSLMs) which can learn the evolutionary landscape of SARS-CoV-2 genomes. By pre-training on over 110 million prokaryotic gene sequences and fine-tuning a SARS-CoV-2-specific model on 1.5 million genomes, we show that GenSLMs can accurately and rapidly identify variants of concern. Thus, to our knowledge, GenSLMs represents one of the first whole genome scale foundation models which can generalize to other prediction tasks. We demonstrate scaling of GenSLMs on GPU-based supercomputers and AI-hardware accelerators utilizing 1.63 Zettaflops in training runs with a sustained performance of 121 PFLOPS in mixed precision and peak of 850 PFLOPS. We present initial scientific insights from examining GenSLMs in tracking evolutionary dynamics of SARS-CoV-2, paving the path to realizing this on large biological data.
ABSTRACT
Surgical tool detection is attracting increasing attention from the medical image analysis community. The goal generally is not to precisely locate tools in images, but rather to indicate which tools are being used by the surgeon at each instant. The main motivation for annotating tool usage is to design efficient solutions for surgical workflow analysis, with potential applications in report generation, surgical training and even real-time decision support. Most existing tool annotation algorithms focus on laparoscopic surgeries. However, with 19 million interventions per year, the most common surgical procedure in the world is cataract surgery. The CATARACTS challenge was organized in 2017 to evaluate tool annotation algorithms in the specific context of cataract surgery. It relies on more than nine hours of videos, from 50 cataract surgeries, in which the presence of 21 surgical tools was manually annotated by two experts. With 14 participating teams, this challenge can be considered a success. As might be expected, the submitted solutions are based on deep learning. This paper thoroughly evaluates these solutions: in particular, the quality of their annotations are compared to that of human interpretations. Next, lessons learnt from the differential analysis of these solutions are discussed. We expect that they will guide the design of efficient surgery monitoring tools in the near future.
Subject(s)
Cataract Extraction/instrumentation , Deep Learning , Surgical Instruments , Algorithms , Humans , Video RecordingABSTRACT
Cardiac tamponade is usually a consequence of increased pericardial pressure with accumulation of pericardial effusion. Pericardial effusion may be caused by acute pericarditis, tumor, uremia, hypothyroidism, trauma, cardiac surgery, or other inflammatory/noninflammatory conditions. In this article we describe four scenarios illustrated by case reports where a small or apparently small pericardial effusion may produce cardiac tamponade. The first scenario illustrates how a small pericardial effusion can cause clinically significant cardiac tamponade when it accumulates rapidly. The second scenario exhibits how an apparently small pericardial effusion on transthoracic echocardiogram (TTE) turned out to be a small amount of unclotted blood and an echogenic hematoma. The third scenario details how an apparently small pericardial effusion on TTE was actually a large loculated effusion in an unusual location seen only by transesophageal echocardiogram (TEE). The fourth scenario demonstrates how the combination of a large pleural effusion and a small pericardial effusion can result in cardiac tamponade. The role of echocardiography in the diagnosis and management of these scenarios is discussed here. Although many clinicians depend on the amount of pericardial effusion to suspect cardiac tamponade, it is important to suspect cardiac tamponade when patients have hemodynamic compromise regardless of the amount of pericardial effusion.
Subject(s)
Cardiac Tamponade/etiology , Pericardial Effusion/complications , Adult , Cardiac Tamponade/diagnosis , Cardiac Tamponade/diagnostic imaging , Cardiac Tamponade/physiopathology , Echocardiography , Echocardiography, Transesophageal , Humans , Male , Middle Aged , Pericardial Effusion/diagnostic imaging , Pericardial Effusion/pathology , Pericardial Effusion/physiopathologySubject(s)
Adrenal Gland Neoplasms/complications , Adrenal Gland Neoplasms/diagnosis , Cardiomyopathies/etiology , Pheochromocytoma/complications , Pheochromocytoma/diagnosis , Pulmonary Edema/etiology , Ventricular Dysfunction, Left/etiology , Humans , Male , Middle Aged , Recovery of Function , Ultrasonography , Ventricular Dysfunction, Left/diagnostic imaging , Ventricular Function, LeftSubject(s)
Coronary Artery Disease/diagnostic imaging , Exercise Test , Heart Ventricles/diagnostic imaging , Posture/physiology , Adult , Aged , Aged, 80 and over , Echocardiography, Stress , Heart Ventricles/physiopathology , Humans , Male , Middle Aged , Ventricular Dysfunction, Left/physiopathologyABSTRACT
A novel ultrasound transducer developed in our laboratory (CONTISON) was used for monitoring catheter balloon commissurotomy (CBC). The transducer was placed at the cardiac apex to obtain an apical four-chamber view and attached to the chest wall using an adhesive ring. During the procedure, the tip of the needle was imaged first in the right atrium and was seen to traverse the interatrial septum and enter the left atrium. Mitral valve gradients were measured before and after CBC.