Researchers in India say they used artificial intelligence (AI) to diagnose mpox using photos of skin lesions with an accuracy up to 99.5%.

Their study, published last week in Medicine in Novel Technology and Devices, tested the ability of the trained deep-learning networks GoogLeNet, Places365-GoogLeNet, SqueezeNet, AlexNet, and ResNet-18 to diagnose mpox using photos of patients' skin lesions (mpox, chickenpox, and other viruses) and computer models designed to mimic human vision.

Currently, mpox is diagnosed using polymerase chain reaction (PCR) testing, but the study authors noted that the results are not always accurate because the virus remains in the blood only a short time. PCR diagnosis also requires information such as rash stage, patient age, and dates of fever or rash onset and is not always available in remote areas.

May aid diagnosis in remote areas

All deep neural networks achieved an accuracy higher than 95%, with ResNet-18 at 99.49%. The researchers noted that ResNet-18's superior performance was likely due to its more straightforward architecture that allows it to learn more complex features with fewer inputs.

"The results prove that deep learning models such as the proposed model based on ResNet-18 can be deployed and can be crucial in battling the [mpox] virus," the authors wrote. "Since the used networks are optimized for efficiency, they can be used on performance limited devices such as smartphones with cameras."

GoogLeNet and Places365-GoogLeNet also perform well for mpox diagnosis and are lightweight and resource-efficient, making them ideal for use in healthcare facilities and in rural areas, they said.

The researchers added that the AI techniques LIME (local interpretable model-agnostic explanations) and GradCAM (gradient-weighted class activation mapping) help health professionals interpret the results to accurately diagnose or rule out mpox.

A global mpox outbreak began in May 2022, mainly affecting men who have sex with men. Transmitted through close contact or contaminated objects, the virus can cause fever, muscle aches, headache, swollen lymph nodes, and other symptoms. As of April 2023, about 87,000 global cases and 112 deaths had been reported.

In a study involving 16 men who have sex with men (MSM) infected with mpox, viral DNA remained detectable in skin lesions for 17 to 31 days after symptom onset. The findings were published in the Journal of the American Academy of Dermatology.

All MSM included in the study had contracted mpox through sexual transmission, and their median age was 36. None were treated with antivirals throughout the course of the study. Swabs were obtained on day 1 of study enrollment and during more than a month of follow-up.

Based on these data, we can question whether three weeks of isolation is sufficient.

On day 1 of the study, all subjects had detectable viral DNA in skin lesions, and 94% had detectable viral DNA in the oropharynx. On day 7, all subjects tested positive in the skin lesions, and 88% tested positive in the oropharynx, the authors said.

"On Day 14, although all subjects had healing skin lesions, most (88%) tested positive in the lesions; in contrast, only 50% tested positive in the oropharynx," the authors said.

The average day of DNA detection in skin lesions following symptom onset was 23 days (95% confidence interval, 21 to 25 days), and 21 days in the oropharynx.

Guidelines might not be sufficient

Currently, the Centers for Disease Control and Prevention (CDC) recommends mpox patients wait 21 days before resuming sexual activity to avoid the spread of the disease.

"Based on these data, we can question whether three weeks of isolation is sufficient," the authors write. "However, the presence of genetic material in samples does not mean the presence of viable viral particles; thus, virus infectivity tests should be conducted to understand the significance of the persistence of viral DNA in skin lesions and the oropharynx."

The incidence of the multidrug-resistant fungal pathogen Candida auris increased dramatically in Israeli hospitals in 2021 and 2022, driven primarily by COVID-19 surges and mechanical ventilation, Israeli researchers reported last week in Emerging Infectious Diseases.

In a nationwide survey, researchers identified 209 patients with C auris infection or colonization from May 2014 to May 2022. Twenty-four of those cases were reported in seven hospitals from May 2014 to December 2020. The C auris incidence rate increased 30-fold in 2021, with an annual incidence of 120 cases reported from 10 hospitals and 3 long-term care facilities. During 2021 and 2022, C auris incidence corresponded with COVID-19 surges in Israel, peaking during the Alpha (January to March 2021), Delta (June to November 2021), and Omicron (January to May 2022) variant waves.

Almost one quarter (23%) of patients with C auris infection or colonization were infected with SARS-CoV-2, and 78% received mechanical ventilation. Analysis of outbreaks at the three hospitals with the most cases indicated C auris spread first among mechanically ventilated COVID-19 patients, then infected ventilated non-COVID-19 patients in intermediate care units and from there spread to non-ventilated patients.

Rise of worrisome clade 3

Multilocus sequence typing revealed that a clade 3 clone, responsible for three cases prior to December 2020, accounted for 55.8% of isolates collected after January 2021. Clade 2 isolates were more frequently resistant to fluconazole (100% vs 63%) and voriconazole (74% vs 5.2%) than non–clade 3 isolates.

The study authors say the use of double-gloving, poor adherence to hand hygiene, and inadequate disinfection of shared medical devices and equipment in COVID-19 units are potential explanations for the C auris outbreaks. They warn that the pathogen appears set to become endemic in some hospitals.

Continued surveillance and implementation of infection control measures, focusing on debilitated patients and those receiving mechanical respiratory support, are essential.

"Continued surveillance and implementation of infection control measures, focusing on debilitated patients and those receiving mechanical respiratory support, are essential to control the spread of C. auris," they wrote.