News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel

News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel
Sign In

Scientists in Italy Develop Hierarchical Artificial Intelligence System to Analyze Bacterial Species in Culture Plates

New artificial intelligence model agrees with interpretations of human medical technologists and microbiologists with extraordinary accuracy

Microbiology laboratories will be interested in news from Brescia University in Italy, where researchers reportedly have developed a deep learning model that can visually identify and analyze bacterial species in culture plates with a high level of agreement with interpretations made by medical technologists.

They initially trained and tested the system to digitally identify pathogens associated with urinary tract infections (UTIs). UTIs are the source for a large volume of clinical laboratory microbiological testing.

The system, known as DeepColony, uses hierarchical artificial intelligence technology. The researchers say hierarchical AI is better suited to complex decision-making than other approaches, such as generative AI.

The researchers published their findings in the journal Nature titled, “Hierarchical AI Enables Global Interpretation of Culture Plates in the Era of Digital Microbiology.”

In their Nature paper, the researchers explained that microbiologists use conventional methods to visually examine culture plates that contain bacterial colonies. The scientists hypothesize which species of bacteria are present, after which they test their hypothesis “by regrowing samples from each colony separately and then employing mass spectroscopy techniques,” to confirm their hypotheses.

However, DeepColony—which was designed for use with clinical laboratory automation systems—looks at high-resolution digital scans of cultured plates and attempts to identify the bacterial strains and analyze them in much the same way a microbiologist would. For example, it can identify species based on their appearance and determine which colonies are suitable for analysis, the researchers explained.

“Working on a large stream of clinical data, and a complete set of 32 pathogens, the proposed system is capable of effectively assisting plate interpretation with a surprising degree of accuracy in the widespread and demanding framework of urinary tract infections,” the study authors wrote. “Moreover, thanks to the rich species-related generated information, DeepColony can be used for developing trustworthy clinical decision support services in laboratory automation ecosystems from local to global scale.”

Alberto Signoroni, PhD

“Compared to the most common solutions based on single convolutional neural networks (CNN), multi-network architectures are attractive in our case because of their ability to fit into contexts where decision-making processes are stratified into a complex structure,” wrote the study’s lead author Alberto Signoroni, PhD (above), Associate Professor of Computer Science, University of Brescia, and his researcher team in their Nature paper. “The system must be designed to generate useful and easily interpretable information and to support expert decisions according to safety-by-design and human-in-the-loop policies, aiming at achieving cost-effectiveness and skill-empowerment respectively.” Microbiologists and clinical laboratory managers will want to follow the further development of this technology. (Photo copyright: University of Brescia.)

How Hierarchical AI Works

Writing in LinkedIn, patent attorney and self-described technology expert David Cain, JD, of Hauptman Ham, LLP, explained that hierarchical AI systems “are structured in layers, each with its own distinct role yet interconnected in a way that forms a cohesive whole. These systems are significant because they mirror the complexity of human decision-making processes, incorporating multiple levels of analysis and action. This multi-tiered approach allows for nuanced problem-solving and decision-making, akin to a seasoned explorer deftly navigating through a multifaceted terrain.”

DeepColony, the researchers wrote, consists of multiple convolutional neural networks (CNNs) that exchange information and cooperate with one another. The system is structured into five levels—labeled 0 through 4—each handling a different part of the analysis:

  • At level 0, the system determines the number of bacterial colonies and their locations on the plate.
  • At level 1, the system identifies “good colonies,” meaning those suitable for further identification and analysis.
  • At level 2, the system assigns each good colony to a bacterial species “based on visual appearance and growth characteristics,” the researchers wrote, referring to the determination as being “pathogen aware, similarity agnostic.”

The CNN used at this stage was trained by using images of 26,213 isolated colonies comprising 32 bacterial species, the researchers wrote in their paper. Most came from clinical laboratories, but some were obtained from the American Type Culture Collection (ATCC), a repository of biological materials and information resources available to researchers.

  • At level 3, the system attempts to improve accuracy by looking at the larger context of the plate. The goal here is to “determine if observed colonies are similar (pure culture) or different (mixed cultures),” the researchers wrote, describing this step as “similarity aware, pathogen agnostic.” This enables the system to recognize variants of the same strain, the researchers noted, and has the effect of reducing the number of strains identified by the system.

At this level, the system uses two “Siamese CNNs,” which were trained with a dataset of 200,000 image pairs.

Then, at level 4, the system “assesses the clinical significance of the entire plate,” the researchers added. Each plate is labeled as:

  • “Positive” (significant bacterial growth),
  • “No significant growth” (negative), or
  • “Contaminated,” meaning it has three or more “different colony morphologies without a particular pathogen that is prevalent over the others,” the researchers wrote.

If a plate is labeled as “positive,” it can be “further evaluated for possible downstream steps,” using MALDI-TOF mass spectrometry or tests to determine susceptibility to antimicrobial measures, the researchers stated.

“This decision-making process takes into account not only the identification results but also adheres to the specific laboratory guidelines to ensure a proper supportive interpretation in the context of use,” the researchers wrote.

Nearly 100% Agreement with Medical Technologists

To gauge DeepColony’s accuracy, the researchers tested it on a dataset of more than 5,000 urine cultures from a US laboratory. They then compared its analyses with those of human medical technologists who had analyzed the same samples.

Agreement was 99.2% for no-growth cultures, 95.6% for positive cultures, and 77.1% for contaminated or mixed growth cultures, the researchers wrote.

The lower agreement for contaminated cultures was due to “a deliberately precautionary behavior, which is related to ‘safety by design’ criteria,” the researchers noted.

Lead study author Alberto Signoroni, PhD, Associate Professor of Computer Science, University of Brescia, wrote in Nature that many of the plates identified by medical technologists as “contaminated” were labeled as “positive” by DeepColony. “We maximized true negatives while allowing for some false positives, so that DeepColony [can] focus on the most relevant or critical cases,” he said.

Will DeepColony replace medical technologists in clinical laboratories any time soon? Not likely. But the Brescia University study indicates the direction AI in healthcare is headed, with high accuracy and increasing speed. The day may not be far off when pathologists and microbiologists regularly employ AI algorithms to diagnose disease.

—Stephen Beale

Related Information:

Hierarchical AI Enables Global Interpretation of Culture Plates in the Era of Digital Microbiology

Hierarchical Deep Learning Neural Network (HiDeNN): An Artificial Intelligence (AI) Framework for Computational Science and Engineering

An AI System Helps Microbiologists Identify Bacteria

This AI Research Helps Microbiologists to Identify Bacteria

Deep Learning Meets Clinical Microbiology: Unveiling DeepColony for Automated Culture Plates Interpretation

Wastewater Analysis Continues to be an Effective Tool for Tracking Deadly Infectious Diseases in Human Communities

In addition to viruses, wastewater analysis can also be used to detect the presence of chemical substances such as opioids

Wastewater surveillance and analysis continues to be a useful tool for detecting the prevalence of viruses such as SARS-CoV-2, influenza, and respiratory syncytial virus (RSV) in a community. Perhaps more importantly, wastewater surveillance can fill in gaps where clinical laboratory testing data may be days or weeks behind the true spread of viral infections.

One sign of the value of testing wastewater for infectious diseases is the fact that government officials are financing a continuing program of wastewater testing. In September, the federal Centers for Disease Control and Prevention (CDC) awarded a contract to conduct wastewater surveillance/analysis worth millions of dollars to Verily Life Sciences, a Google company, rather than renewing its contract with Biobot Analytics, which had been doing the work since 2020. One interesting twist in the award of this contract is how an ensuing dispute pulled the plug on a significant portion of the wastewater analysis in this country.

In their September Morbidity and Mortality Weekly Report (MMWR), the CDC highlighted a CDC study during which wastewater samples were taken from 40 wastewater treatment plants located in Wisconsin’s three largest cities. The samples were collected weekly and tested for influenza and RSV. The findings were then compared with data regarding emergency department (ED) visits for those diseases.

The CDC found that higher detections of flu and RSV were associated with higher rates of ED visits for both illnesses. The study also suggests that wastewater might detect the spread of these viruses earlier than ED visit data alone.

Peter DeJonge, PhD

“During the COVID-19 pandemic, wastewater surveillance for SARS-CoV-2 provided valuable insight into community incidence of COVID-19,” said Peter DeJonge, PhD (above), a CDC Career Epidemiology Field Officer, in an interview with Infectious Disease Special Edition. “[The CDC’s] report supports the idea that wastewater surveillance also has the potential to serve as a useful method with which to track community spread of influenza and RSV.” Local clinical laboratories are also involved in the CDC’s wastewater surveillance programs. (Photo copyright: CDC.)


Keeping Communities Informed about Spread of Viral Infections

The CDC’s study was conducted from August 2022 to March 2023. The wastewater samples from all three cities tested positive for the viruses in advance of increases in ED visits. After the ED visits for those viruses had subsided, the viral material remained in sewersheds for up to three months. 

“Both influenza and RSV can cause substantial amounts of illness, hospitalization, and even death during annual epidemics, which often occur during winter months in the US,” Peter DeJonge, PhD, a CDC Career Epidemiology Field Officer assigned to the Chicago Department of Public Health, told Infectious Disease Special Edition (IDSE). “Clinical providers and public health officials benefit from surveillance data to understand when and where these diseases are spreading in a community each year. This type of data can help prepare clinics [and clinical laboratories] for anticipated cases, tailor public health messaging, and encourage timely vaccination.”

“The collective burden from these respiratory viruses is staggering. With these viruses circulating simultaneously and potentially shifting in seasonality and severity, communities must be able to understand the full impact of each of these illnesses to inform awareness and public health responses that can prevent infections, hospitalizations, and even deaths,” said Mariana Matus, PhD, CEO and cofounder of Biobot Analytics, in an August press release announcing the launch of a “Respiratory Illnesses Panel” that will monitor wastewater for Influenzas A and B (seasonal flu), Respiratory Syncytial Virus (RSV), and SARS-CoV-2 (COVID-19).

“Traditional testing methods for these illnesses do not provide a comprehensive picture of the number of people infected due to inaccurate reporting, as well as asymptomatic or misdiagnosed cases,” Matus continued. “By monitoring wastewater concurrently for influenza, RSV, and SARS-CoV-2, we can fill in these gaps and provide important information to communities.”

CDC Moves to Change Wastewater Surveillance Contractor Mid-stream

As new variants of SARS-CoV-2 emerge, a recent contract dispute may be the cause of a time delay in efforts to perform wastewater surveillance for the disease, as well as for other viral infections, according to Politico.

The CDC’s move to replace Biobot Analytics with Verily Life Sciences to do wastewater surveillance has led to Biobot filing a protest with the Government Accountability Office (GAO).

According to World Socialist Web Site (WSWS), “The scope of the [Biobot] contract [to provide extended data for the public health agency’s National Wastewater Surveillance System (NWSS)] included data from more than 400 locations from over 250 counties across the entire United States, covering 60 million people. On top of this, Biobot also conducted genomic sequencing to identify the latest variants in circulation.” 

About one quarter of the wastewater testing sites in the country have been shut down due to Biobot’s contract being suspended in September. The remaining 1,200 sites that are not covered under the original contract will continue wastewater testing, Politico reported. 

The GAO hopes to have a decision on the contract dispute in January. Verily says it is ready to proceed with testing in all locations and already has its infrastructure in place. 

“We are committed to working with the CDC to advance the goals of the … testing program, initiate testing on the samples already delivered when allowed to resume work, and make wastewater data available as quickly as possible,” Bradley White, PhD, Principal Scientist/Director at Verily, told Politico.

Under the terms of Verily’s contract, the company will collect samples from wastewater treatment centers cross the county and analyze the samples for COVID-19 and the mpox (monkey pox) virus.

This contract marks the first agreement between the CDC and Verily.

The CDC has not disclosed why it decided to change contractors, but it is probable that cost may have been played a role in the decision. Verily’s contract is for $38 million over the course of five years and Biobot’s most recent contract was for around $31 million for a period of less than 18 months, Politico reported. 

In a LinkedIn post, Matus reported that Biobot had already laid off 35% of its staff due to the contract decision by the CDC. 

Competition in Wastewater Surveillance Market

When seeking viruses in wastewater, scientists use gene-based detection methods to locate and amplify genetic signs of pathogens. But public health officials are just beginning to tap into the potential opportunities that may exist in the analysis of data present in wastewater.

Wastewater surveillance is also being looked at as a way to combat America’s opioid epidemic.

“Wastewater surveillance is becoming more mature and more mainstream month after month, year over year,” Matus told Time

Thus, regardless of which companies end up working with the CDC, it appears that wastewater surveillance and analysis, which requires a great deal of clinical laboratory testing, will continue to help fight the spread of deadly viral infections, as well as possibly the nation’s drug epidemic.

—JP Schlingman

Related Information:

Wastewater Shows COVID Levels Dipping as Hospitalizations Tick Up

How Wastewater Testing Can Help Tackle America’s Opioid Crisis

Wastewater Surveillance May Help Detect Flu, RSV Outbreaks

The Respiratory Illnesses Panel Will Include Monitoring for Influenza A and B, RSV, and SARS-CoV-2

Wastewater Surveillance Data as a Complement to Emergency Department Visit Data for Tracking Incidence of Influenza A and Respiratory Syncytial Virus—Wisconsin, August 2022–March 2023

Biobot Analytics Files Protest against CDC Issuing Wastewater Surveillance Contract to Verily

Biobot Analytics Awarded NIDA Funding for Nationwide Wastewater-based Monitoring Program for High Risk Substance and Others Associated with Health Risks

Wastewater Signals Upswing in Flu, RSV

Biobot Analytics Launches Respiratory Illness Panel

Detecting COVID Surges is Getting Harder, Thanks to a Contract Dispute

Verily Lands $38 Million Deal with CDC for Wastewater Surveillance

Genetic Testing of Wastewater Now Common in Detecting New Strains of COVID-19 and Other Infectious Diseases

San Francisco International Airport First in the Nation to Test Wastewater for SARS-CoV-2 Coronavirus

;