Automated Interpretation of Blood Culture Gram Stains by Use of a Deep Convolutional Neural Network

Using Artificial Intelligence in Infection Prevention

Purpose of Review

Artificial intelligence (AI) offers huge potential in infection prevention and control (IPC). We explore its potential IPC benefits in epidemiology, laboratory infection diagnosis, and hand hygiene.

Recent Findings

AI has the potential to detect transmission events during outbreaks or predict high-risk patients, enabling development of tailored IPC interventions. AI offers opportunities to enhance diagnostics with objective pattern recognition, standardize the diagnosis of infections with IPC implications, and facilitate the dissemination of IPC expertise. AI hand hygiene applications can deliver behavior change, though it requires further evaluation in different clinical settings. However, staff can become dependent on automatic reminders, and performance returns to baseline if feedback is removed.

Summary

Advantages for IPC include speed, consistency, and capability of handling infinitely large datasets. However, many challenges remain; improving the availability of high-quality representative datasets and consideration of biases within preexisting databases are important challenges for future developments. AI in itself will not improve IPC; this requires culture and behavior change. Most studies to date assess performance retrospectively so there is a need for prospective evaluation in the real-life, often chaotic, clinical setting. Close collaboration with IPC experts to interpret outputs and ensure clinical relevance is essential.

Consolidation of Clinical Microbiology Laboratories and Introduction of Transformative Technologies

Clinical microbiology is experiencing revolutionary advances in the deployment of molecular, genome sequencing-based, and mass spectrometry-driven detection, identification, and characterization assays. Laboratory automation and the linkage of information systems for big(ger) data management, including artificial intelligence (AI) approaches, also are being introduced. The initial optimism associated with these developments has now entered a more reality-driven phase of reflection on the significant challenges, complexities, and health care benefits posed by these innovations. With this in mind, the ongoing process of clinical laboratory consolidation, covering large geographical regions, represents an opportunity for the efficient and cost-effective introduction of new laboratory technologies and improvements in translational research and development. This will further define and generate the mandatory infrastructure used in validation and implementation of newer high-throughput diagnostic approaches. Effective, structured access to large numbers of well-documented biobanked biological materials from networked laboratories will release countless opportunities for clinical and scientific infectious disease research and will generate positive health care impacts. We describe why consolidation of clinical microbiology laboratories will generate quality benefits for many, if not most, aspects of the services separate institutions already provided individually. We also define the important role of innovative and large-scale diagnostic platforms. Such platforms lend themselves particularly well to computational (AI)-driven genomics and bioinformatics applications. These and other diagnostic innovations will allow for better infectious disease detection, surveillance, and prevention with novel translational research and optimized (diagnostic) product and service development opportunities as key results.

Machine learning applications in epilepsy

Machine learning leverages statistical and computer science principles to develop algorithms capable of improving performance through interpretation of data rather than through explicit instructions. Alongside widespread use in image recognition, language processing, and data mining, machine learning techniques have received increasing attention in medical applications, ranging from automated imaging analysis to disease forecasting. This review examines the parallel progress made in epilepsy, highlighting applications in automated seizure detection from electroencephalography (EEG), video, and kinetic data, automated imaging analysis and pre-surgical planning, prediction of medication response, and prediction of medical and surgical outcomes using a wide variety of data sources. A brief overview of commonly used machine learning approaches, as well as challenges in further application of machine learning techniques in epilepsy, is also presented. With increasing computational capabilities, availability of effective machine learning algorithms, and accumulation of larger datasets, clinicians and researchers will increasingly benefit from familiarity with these techniques and the significant progress already made in their application in epilepsy.

Artificial Intelligence in Medicine: Weighing the Accomplishments, Hype, and Promise

INTRODUCTION

Artificial Intelligence in Medicine (AIM) research is now 50 years old, having made great progress that has tracked the corresponding evolution of computer science, hardware technology, communications, and biomedicine. Characterized as being in its "adolescence" at an international meeting in 1991, and as "coming of age" at another meeting in 2007, the AIM field is now more visible and influential than ever before, paralleling the enthusiasm and accomplishments of artificial intelligence (AI) more generally.

OBJECTIVES

This article summarizes some of that AIM history, providing an update on the status of the field as it enters its second half-century. It acknowledges the failure of AI, including AIM, to live up to early predictions of its likely capabilities and impact.

METHODS

The paper reviews and assesses the early history of the AIM field, referring to the conclusions of papers based on the meetings in 1991 and 2007, and analyzing the subsequent evolution of AIM.

CONCLUSION

We must be cautious in assessing the speed at which further progress will be made, despite today's wild predictions in the press and large investments by industry, including in health care. The inherent complexity of medicine and of clinical care necessitates that we address issues of usability, workflow, transparency, safety, and formal clinical trials. These requirements contribute to an ongoing research agenda that means academic AIM research will continue to be vibrant while having new opportunities for more interactions with industry.

Automated classification of osteomeatal complex inflammation on computed tomography using convolutional neural networks

BACKGROUND

Convolutional neural networks (CNNs) are advanced artificial intelligence algorithms well suited to image classification tasks with variable features. These have been used to great effect in various real-world applications including handwriting recognition, face detection, image search, and fraud prevention. We sought to retrain a robust CNN with coronal computed tomography (CT) images to classify osteomeatal complex (OMC) occlusion and assess the performance of this technology with rhinologic data.

METHODS

The Google Inception-V3 CNN trained with 1.28 million images was used as the base model. Preoperative coronal sections through the OMC were obtained from 239 patients enrolled in 2 prospective chronic rhinosinusitis (CRS) outcomes studies, labeled according to OMC status, and mirrored to obtain a set of 956 images. Using this data, the classification layer of Inception-V3 was retrained in Python using a transfer learning method to adapt the CNN to the task of interpreting sinonasal CT images.

RESULTS

The retrained neural network achieved 85% classification accuracy for OMC occlusion, with a 95% confidence interval for algorithm accuracy of 78% to 92%. Receiver operating characteristic (ROC) curve analysis on the test set confirmed good classification ability of the CNN with an area under the ROC curve (AUC) of 0.87, significantly different than both random guessing and a dominant classifier that predicts the most common class (p < 0.0001).

CONCLUSION

Current state-of-the-art CNNs may be able to learn clinically relevant information from 2-dimensional sinonasal CT images with minimal supervision. Future work will extend this approach to 3-dimensional images in order to further refine this technology for possible clinical applications.

Laboratory Automation in Clinical Microbiology

Laboratory automation is currently the main organizational challenge for microbiologists. Automating classic workflows is a strenuous process for the laboratory personnel and a huge and long-lasting financial investment. The investments are rewarded through increases in quality and shortened time to report. However, the benefits for an individual laboratory can only be estimated after the implementation and depending on the classic workflows currently performed. The two main components of automation are hardware and workflow. This review focusses on the workflow aspects of automation and describes some of the main developments during recent years. Additionally, it tries to define some terms which are related to automation and specifies some developments which would further improve automated systems.

Waltzing around Sacred Cows on the Way to the Future

Our mostly manual, agar-based clinical microbiology laboratory is slowly but steadily being redefined by automation and innovation. Ironically, the oldest test, the Gram stain test, is still manually read and interpreted by trained personnel. In a proof-of-concept study, Smith et al. (J. Clin. Microbiol. 56:e01521-17, 2018, https://doi.org/10.1128/JCM.01521-17) used computer imaging with a deep convolutional neural network to examine and interpret Gram-stained slides from positive blood culture bottles. In light of the shortage of medical technologists/microbiologists and the need for results from positive blood culture bottles 24/7, this paper paves the way for the next innovations for the clinical microbiology laboratory of the future.