Artificial Intelligence-Based Screening for Mycobacteria in Whole-Slide Images of Tissue Samples

OBJECTIVES

This study aimed to develop and validate a deep learning algorithm to screen digitized acid fast-stained (AFS) slides for mycobacteria within tissue sections.

METHODS

A total of 441 whole-slide images (WSIs) of AFS tissue material were used to develop a deep learning algorithm. Regions of interest with possible acid-fast bacilli (AFBs) were displayed in a web-based gallery format alongside corresponding WSIs for pathologist review. Artificial intelligence (AI)-assisted analysis of another 138 AFS slides was compared to manual light microscopy and WSI evaluation without AI support.

RESULTS

Algorithm performance showed an area under the curve of 0.960 at the image patch level. More AI-assisted reviews identified AFBs than manual microscopy or WSI examination (P < .001). Sensitivity, negative predictive value, and accuracy were highest for AI-assisted reviews. AI-assisted reviews also had the highest rate of matching the original sign-out diagnosis, were less time-consuming, and were much easier for pathologists to perform (P < .001).

CONCLUSIONS

This study reports the successful development and clinical validation of an AI-based digital pathology system to screen for AFBs in anatomic pathology material. AI assistance proved to be more sensitive and accurate, took pathologists less time to screen cases, and was easier to use than either manual microscopy or viewing WSIs.

A perspective on intelligent devices and environments in medical rehabilitation

Globally, the number of people older than 65 years is anticipated to double between 1997 and 2025, while at the same time the number of people with disabilities is growing at a similar rate, which makes technical advances and social policies critical to attain, prolong, and preserve quality of life. Recent advancements in technology, including computation, robotics, machine learning, communication, and miniaturization of sensors have been used primarily in manufacturing, military, space exploration, and entertainment. However, few efforts have been made to utilize these technologies to enhance the quality of life of people with disabilities. This article offers a perspective of future development in seven emerging areas: translation of research into clinical practice, pervasive assistive technology, cognitive assistive technologies, rehabilitation monitoring and coaching technologies, robotic assisted therapy, and personal mobility and manipulation technology.

A prototype power assist wheelchair that provides for obstacle detection and avoidance for those with visual impairments

Background

Almost 10% of all individuals who are legally blind also have a mobility impairment. The majority of these individuals are dependent on others for mobility. The Smart Power Assistance Module (SPAM) for manual wheelchairs is being developed to provide independent mobility for this population.

Methods

A prototype of the SPAM has been developed using Yamaha JWII power assist hubs, sonar and infrared rangefinders, and a microprocessor. The prototype limits the user to moving straight forward, straight backward, or turning in place, and increases the resistance of the wheels based on the proximity of obstacles. The result is haptic feedback to the user regarding the environment surrounding the wheelchair.

Results

The prototype has been evaluated with four blindfolded able-bodied users and one individual who is blind but not mobility impaired. For all individuals, the prototype reduced the number of collisions on a simple navigation task.

Conclusion

The prototype demonstrates the feasibility of providing navigation assistance to manual wheelchair users, but several shortcomings of the system were identified to be addressed in a second generation prototype.