Daily Archives: October 16, 2023

DES MOINES More than $11 billion in federal funding was sent to Iowa via multiple aid packages designed to help Americans and local governments weather the effects of the COVID-19 pandemic.

More than $2.4 billion remains unspent by Iowa, according to a recent state report. But how was the rest of it spent? And whats planned for the remaining funds? Those questions were recently posed to The Gazette by a curious Iowan.

Curious Iowa is a series from The Gazette that seeks to answer Iowans questions about the state, its culture and the people who live here. To answer that question, we took a look at a recent report from the nonpartisan Iowa Legislative Services Agency. It analyzed how much the state received in federal financial assistance during the pandemic, where the state has allocated funds, and how much it still has to spend.

LSA issued the report in early August, and the financial analysis within the report is accurate as of July 28. That means the figures in the report are likely slightly different as of mid-October. The report tracks funding from 87 different federal programs.

State agencies in Iowa have reported receiving federal assistance totaling $11.047 billion, according to the LSA report.

That federal assistance came from six pieces of federal funding legislation passed in 2020 and 2021, including the CARES Act and Paycheck Protection Program passed in 2020 under former President Donald Trump, and the American Rescue Plan Act passed in 2021 under President Joe Biden.

Broadly speaking, $7.9 billion was allocated directly to state agencies, nearly $3 billion was allocated to the states Coronavirus Fiscal Recovery Fund, and $152 million was dedicated to a capital fund for broadband internet expansion.

Broken down by program, Iowas unemployment insurance program received the largest chunk of federal financial assistance: nearly $3 billion, according to the LSA report.

During the peak of the pandemic, unemployment claims spiked in Iowa and across the country as businesses were forced to close either by public health mandate or because of a lack of business. A large portion of the federal financial assistance was designed to help employers and workers survive that disruption.

More than $1.1 billion went to Iowas Medical Assistance Program, which supports the states Medicaid program.

Roughly $650 million in federal assistance designed to help state and local governments was sent to Iowa.

And a trio of education programs each received roughly $400 million in federal funds.

Given those programs received the most pandemic assistance funding, it comes as no surprise that the three state agencies that received the most funding were Workforce Development with $3 billion, Health and Human Services with $1.6 billion, and Education with $1.1 billion.

Of that $2.4 billion that the state has not yet allocated, the biggest chunks are in the state public health department and the Iowa Finance Authority, which operates housing assistance programs for homeowners and renters and has spent less than a third of its allotted federal funding.

Most of the funds have deadlines by which the money must be spent, and those deadlines are staggered over the coming years, depending on the funding source. In some cases, the state has until 2027 to spend pandemic relief funds.

There is nearly $1.1 billion remaining in the Iowa Coronavirus Fiscal Recovery Fund, according to the LSA report. Those funds carry broad spending authority: they can be used to address emergency and economic effects of the pandemic, replenish lost government revenues, or invest in water, sewer and broadband infrastructure.

Funds from the fiscal recovery fund must be spent by the end of 2026, or they will revert back to the federal government.

The public health department, which is now under the Health and Human Services department with the recent reorganization of state government, is yet to spend more than $337 million, according to the LSA report. Thats 60 percent of the federal funds awarded to the department.

The majority of the public health departments unspent funds are for the Epidemiology and Laboratory Capacity for Infectious Diseases Program, according to the LSA report. Expenses for that program are ongoing and funding will remain available in future years, the report says.

The Iowa Finance Authority is yet to spend $208 million; thats 68 percent of its allotted financial assistance. The bulk of those unspent funds are for the Emergency Rental Assistance program, and those funds also will remain available in future years, according to the LSA report.

The workforce development agency had $415.5 million still unspent, but that is just 12 percent of the funding it received.

Tell us what youd like us to investigate next.

Comments: (515) 355-1300, erin.murphy@thegazette.com

Go here to see the original:

Curious Iowa: Has the state spent all of its COVID-19 relief funding? - The Gazette

IA2030 has seven strategic priorities: integration of immunization with primary care, country commitment to immunization fulfilling people-centered demand for vaccines, ensuring high and equitable coverage, vaccinating throughout the life course and integrated with essential services, management of outbreaks and emergencies, sustainable supplies of vaccines, and research and innovation. The COVID-19 vaccination campaign was well aligned with all seven of IA2030s strategic priorities. The priorities that can impart the most momentum to the program and support the global IA2030 vision are ones that increase the number of vaccines in the routine program (Strategic Priority 2, Commitment and Demand) and broaden the age groups recommended for routine vaccination (Strategic Priority 4, Life Course & Integration). Another strategic priority with long-term benefit for China and the world is Strategic Priority 7, Research & Innovation. The COVID-19 campaign relied heavily on research and innovation, as exemplified by rapid and successful development of new and innovative COVID-19 vaccines, new vaccine delivery techniques, and new vaccination strategies in a fully digitalized immunization program that harnessed big data and artificial intelligence for monitoring and analyzing vaccine safety, effectiveness, and coverage.

IA2030 talks about breadth of protection, meaning implementing and sustaining high coverage with all WHO-recommended vaccines. The COVID-19 vaccination campaign conclusively proved that Chinas immunization program can rapidly develop and introduce a new vaccine and achieve high vaccine coverage. The COVID-19 vaccination campaign was an astonishing eight times the size of the annual routine immunization program. This momentum and experience can be used to facilitate introduction of the vaccines that are recommended by WHO for all national programs but are not currently in Chinas program. Recent [8, 9] and earlier [10] analyses of Chinas national immunization program have recognized the disparity between the vaccines recommended by WHO and the vaccines included in Chinas program and have recommended strategies to introduce new vaccines. A legislatively supported mechanism, the National Immunization Advisory Committee (NIAC), now exists that can recommend to government non-program vaccines that should be moved into the program based on evidence of preventable burden of disease and vaccine effectiveness, safety, cost effectiveness, and supply security [11]. Just as NIAC supported COVID-19 vaccination strategy, it can support evidence-based introduction of other vaccines. Moving human papillomavirus (HPV), pneumococcal conjugate (PCV), influenza, Haemophilus influenzae type b (Hib), varicella, and rotavirus vaccines into the national program would bring equitable and high coverage of these vaccines to well over 100 million young children and adolescents, preventing suffering from these infectious diseases while saving society money. Using domestically developed and produced vaccines will strengthen Chinas vaccine industry and foster innovative development of new vaccines for use in China and for WHO prequalification and global use. For example, combining Chinas Sabin-strain inactivated poliovirus vaccines into diphtheria, tetanus, acellular pertussis-, Hib-, and hepatitis B-containing combination vaccines could make space in the domestic routine immunization schedule for other vaccines while maintaining high polio vaccine coverage well into the future, for as long as is needed in China and elsewhere [12].

A thrust of IA2030 is life course vaccination. With its target population of everyone over the age of 3years, Chinas COVID-19 vaccination campaign exemplified life-course vaccination. COVID-19 vaccination was vigorously promoted to the elderly [13], people with comorbidities, health care workers and other working age adults, and school-age children. These are the same target populations for seasonal influenza vaccine. China CDC has recommended influenza vaccination of these populations for years [14], however uptake has been low except in several leading cities that have embraced influenza vaccination of these key target populations. The COVID-19 vaccination campaign proved that these populations can be reached to achieve high and equitable coverage. The COVID-19 vaccination experience can be used to make progress on influenza vaccination of these important target populations.

The maximum age for eligibility of National Immunization Program vaccines was recently raised from 14 to 18years of age. This age range expansion can provide adolescents the opportunity to catch up on any program vaccinations they missed and to receive HPV vaccine once it is included in the program. But why stop at 18years of age? The IA2030 vision is for the entire life course, as was the COVID-19 campaign. Program eligibility across all ages would enable immunization clinics, community health centers, and primary care providers to bring "everyone, everywhere, at all ages the full benefits of vaccines. For adults, this could include not only influenza vaccine, but also pneumococcal and zoster vaccines in a comprehensive program integrated with primary care. Innovative financing of adult vaccines, as was done for COVID-19 vaccines with the Medical Insurance Fund, could support universal immunization program eligibility that would lead to equitable and high coverage for allin good alignment with IA2030.

Here is the original post:

Accelerating into Immunization Agenda 2030 with momentum from ... - Infectious Diseases of Poverty - BioMed Central

Register for free to listen to this article

Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

New research helps explain why young children have lower rates of severe COVID-19 than adults. A study of infants and young children found those who acquired SARS-CoV-2 had a strong, sustained antibody response to the virus and high levels of inflammatory proteins in the nose but not in the blood. This immune response contrasts with that typically seen in adults with SARS-CoV-2 infection. Co-funded by the National Institute of Allergy and Infectious Diseases (NIAID), part of the National Institutes of Health, the research appears in the journal Cell.

The investigation involved 81 full-term infants and young children whose mothers enrolled in a NIAID-supported cohort study at Cincinnati Childrens during their third trimester of pregnancy. The study team trained mothers to collect weekly nasal swabs from their infants starting when the babies were 2 weeks old. The team also drew blood from the babies regularly, starting at age 6 weeks, as well as when the children became infected with SARS-CoV-2 and during subsequent weeks and months.

Subscribe to Technology Networks daily newsletter, delivering breaking science news straight to your inbox every day.

These samples enabled the scientists to study the childrens immune responses before, during and after they were exposed to the virus for the first time. Fifty-four of the children became infected and had mild COVID-19, while 27 who tested negative through the study period served as matched controls. At the time of infection, the children were 1 month to nearly 4 years old, and half were 9 months or younger. The study also included weekly nasal swabs from 19 mothers with COVID-19 and 19 healthy mothers as controls, as well as blood samples from 89 adults with COVID-19 and 13 healthy controls.

The researchers examined many aspects of the babies and adults immune responses to the virus through an approach calledsystems immunology. The study revealed that young childrens antibody response to SARS-CoV-2 differs from that of adults. Typically, adults produce antibodies to the virus at levels that spike for a few weeks, then decline. In contrast, the infants and young children in the study produced protective antibodies at levels that spiked and remained high for up to the full 300-day observation period.

The scientists also found that the blood of adults with SARS-CoV-2 infection typically had high levels of proteins called inflammatory cytokines, which are associated with severe COVID-19 and death, while the blood of babies and children did not. However, the childrens noses had high levels of inflammatory cytokines and a potent antiviral cytokine.

According to the researchers, these findings suggest that cytokines snuffed out SARS-CoV-2 infection right at the site where the virus entered the childrens bodies, potentially explaining the mildness of their COVID-19 disease. The findings also suggest it may be possible to devisevaccine adjuvantsthat mimic the immune responses observed in young children by stimulating persistently high antibody levels without causing dangerous excess inflammation in the blood.

Children aged 6 months to 4 years who got COVID-19 vaccines before September 12, 2023, should get one or two doses of updated COVID-19 vaccine, depending on which vaccine and how many doses they previously received. Children aged 6 months to 4 years who have not been vaccinated should get two or three doses of updated COVID-19 vaccine, depending on which vaccine they receive.

Reference:Wimmers F, Burrell AR, Feng Y, et al. Multi-omics analysis of mucosal and systemic immunity to SARS-CoV-2 after birth. Cell. 2023;186(21):4632-4651.e23. doi:10.1016/j.cell.2023.08.044

This article has been republished from the following materials. Note: material may have been edited for length and content. For further information, please contact the cited source.

More:

Study Uncovers Why Young Children Suffer Less Severe COVID-19 - Technology Networks

WASHINGTON Select Subcommittee on the Coronavirus Pandemic Chairman Brad Wenstrup (R-Ohio) has joined forces with all Majority Members to shed light on coercive and potentially harmful COVID-19 policies that are reemerging at the University of Maryland. Under the Universitys new directive, Maryland students who test positive for COVID-19 are to be immediately removed from their dorms and forced into isolation, either at a nearby hotel or by boarding a flight home presumably at their own expense. Considering the University of Maryland received hundreds of millions of dollars in COVID-19 aid, this new directive raises questions about how the University utilized its federal relief funds and how it plans to enforce this highly destructive directive.

Maryland seems to be reinstituting the same negative policies it implemented during the beginning of the pandemic at the expense of its students. We know more now, according to public reports, Maryland is removing students who test positive for COVID-19 from their dorms without providing temporary housing accommodations and sending them to their permanent homeslikely with their older, more at-risk, parents.In other cases, studentsas mandated by the Directiveare required to isolate at a nearby hotel. Presumably, its the students parentsnot your universitythat are footing the bill, which begs the question of how Maryland spent the federal Coronavirus dollars it received, wrote Chairman Wenstrup.

The Select Subcommittees investigation previously revealed that prolonged school closures and hybrid learning programs contributed to a mental health crisis, historic learning loss, and decreased physical health among Americas youth. Reinstituting policies that will have a similar effect on students is not only counterproductive, but also proves nonsensical when considering available science and data.

Read the Select Subcommittees full letter to University of Maryland President Darryll Pines here.

###

Link:

Wenstrup, Select Subcommittee Majority Members Investigate ... - House Committee on Oversight and Reform |

When Connecticut saw a jump in COVID-19 cases in September, Abby M. Warren, a partner at law firm Robinson+Cole, started to get a trickle of calls from employers interested in policy recommendations for staff who get exposed to the virus.

Warren, an employment lawyer, is expecting those inquiries to increase in the coming weeks and months, as colder, drier weather and holiday gatherings bring on the traditional respiratory virus season and a rise in COVID infections.

With state and federal emergency COVID-19 declarations in the rearview mirror they were lifted in the spring Warren and other experts say not only are mask and vaccination mandates out, but so are workplace policies seen as intruding on personal liberties, such as mandatory temperature checks and repeated questioning about potential exposure.

As a replacement, employment lawyers and many companies are recommending policies that require common-sense precautions, like telling workers to avoid the workplace when ill.

Now is an especially good time of year for employers to update their illness policies and remind employees of expectations, Warren said.

Even if there is not a policy, employers should be sending out a message about cold and flu season and their expectations, Warren said. Or else, someone gets a cold, and they dont know. Or, they assume its just a cold and they come in.

Under the pandemic emergency declarations, employers could more easily perform temperature checks and regularly ask COVID-19 exposure screening questions, Warren said.

And now we are back to the standard where it has to be supported by business necessity because that is a medical exam and medical exams are regulated under the Americans with Disabilities Act and a few other laws, Warren said.

Susan Huntington, chair of law firm Day Pitneys healthcare practice and a certified physicians assistant, said her clients, largely healthcare providers, are no longer requiring follow-up COVID-19 booster shots.

Thats a significant change compared to January 2022, when Gov. Ned Lamont signed an executive order that required employees of all long-term care facilities and state hospitals in Connecticut to receive COVID-19 boosters. The Connecticut Hospital Association at that time also implemented a similar mandatory booster shot requirement for all hospital and health system employees.

Now, Huntingtons clients are predominantly concerned about how to react when an employee tests positive, or exhibits COVID-19 symptoms.

Huntington said her policy recommendations are based on the latest guidance from the U.S. Centers for Disease Control and Prevention.

The CDC recommends people who test positive for COVID-19 isolate at home for at least five days and then wear a high-quality mask when in public for an additional five days. Huntington also recommends employers require masks for staff with respiratory ailment symptoms, even if they test negative for COVID.

Employers still have an obligation to provide a safe work environment under (Occupational Safety and Health Administration) rules, Huntington said. And you wouldnt think you would have to tell someone not to come into the office if theyre coughing, sneezing or have stomach-bug symptoms. But people do.

New Britain-based manufacturer Stanley Black & Decker, which has 50,000 employees worldwide, launched an internal campaign this summer aimed at encouraging employees to take more direct control over COVID precautions through common-sense actions.

Stanleys Our People, Our Health campaign used posters and other communication to urge employees to stay home when sick and take other measures recommended by the CDC to keep colleagues safe.

There could be incentives to come to work sick, and we didnt want to have them, said Dr. Mitch McClure, Stanleys chief medical officer.

The idea is also to step away from mandates and intrusive questions about COVID testing, travel and other topics that arent as justifiable now as they were before the end of the emergency declaration, McClure said.

Now, the onus is really on you to self-identify because we dont want to be inappropriate, McClure said. We dont want to be asking people about their personal health information because thats not our place once the (emergency) authorization has ended.

Drew Andrews, managing partner and CEO of Hartford-based accounting and consulting firm Whittlesey, said he sees little need to update COVID policies because most of his roughly 150 employees continue to opt to work remotely.

I havent had to deal with it because I have to go looking hard to find people, Andrews joked. Its like social distancing is here because I dont have anyone here. There is no reason to add extra layers of compliance to people and give them a hard time.

Andrews said he has, however, continued policies implemented during the pandemic, including a requirement for sick workers to remain at home. That is made easier by another recent company mandate that all staff take their laptop computers home every day.

Andrews said there have been several recent COVID cases among his staff, but all have reported milder symptoms. The virus doesnt seem to raise much alarm anymore among Andrews colleagues and the companies they serve.

I think its almost becoming like the seasonal flu or cold and I see that across the board and in companies, Andrews said.

During an Oct. 5 visit to the Fair Haven Community Health Care Clinic in New Haven, Connecticut Department of Public Health Commissioner Dr. Manisha Juthani and Mandy K. Cohen, director of the U.S. Centers for Disease Control and Prevention, acknowledged reduced risks of the COVID-19 virus but stressed a continuing urgency for vaccination and other protections.

PHOTO | SHAHRZAD RASEKH/CT MIRROR

Centers for Disease Control Director Dr. Mandy Cohen and Rep. Rosa DeLauro visited Fair Haven Community Health Care and answered questions from the press about COVID-19 risks and vaccinations.

Healthcare authorities say they expect COVID hospitalizations to be on par with last years fall and winter season, enough to strain healthcare providers.

The latest COVID vaccine is free to all Americans, either through private health insurance or coverage from the federal government.

Cohen noted that 97% of the population has some degree of immunity, either through vaccination or prior infection. But that protection wanes over time, she said.

Cohen said a late-summer rise in COVID cases has abated, but she anticipates a seasonal increase.

We know as we get into the fall and winter, where we are doing activities where this virus likes to spread, we fully expect to see this virus continue to circulate, continue to go up, just like weve seen this last winter and fall, Cohen said.

Juthani acknowledged there have been some initial vaccine shortages as private providers take over distribution from government agencies. She expressed guarded optimism the vaccine will be reliably available throughout Connecticut by mid-October.

Juthani said employers should require ill staff to remain at home. They can also guard against COVID spread with state-of-the-art ventilation systems. Beyond that, Juthani said the current situation doesnt call for prior measures like universal masking, spacing desks six feet apart or staggering employee attendance.

I dont think the original precautions that we had are necessary at this stage of this virus and where we are, Juthani said.

Yale University epidemiologist and professor Dr. Sten H. Vermund said between those who have been vaccinated and those who have recovered from past infection, there is only a tiny subset of the population susceptible to significant risk of serious health consequences from COVID-19.

He said strict employer health mandates would likely be ignored.

Instead, companies can reduce infection risk with upgraded ventilation systems that include better filtration. Employers should also strongly encourage and facilitate vaccinations, offering time off for appointments or arranging workplace vaccination clinics, Vermund said.

Read more:

With COVID-19 emergency orders lifted, employers seek guidance ... - Hartford Business Journal

This retrospective study was approved by the institutional review boards of eight hospitals (Kobe University Hospital, St. Luke's International Hospital, Nishinomiya Watanabe Hospital, Kobe City Medical Center General Hospital, Kobe City Nishi-Kobe Medical Center, Hyogo Prefectural Kakogawa Medical Center, Kita Harima Medical Center, and Hyogo Prefectural Awaji Medical Center); the requirement for acquiring informed consent was waived by the institutional review boards of these eight hospitals owing to the retrospective nature of the study. This study complied with the Declaration of Helsinki and Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan (https://www.mhlw.go.jp/file/06-Seisakujouhou-10600000-Daijinkanboukouseikagakuka/0000080278.pdf).

The CXR datasets used for developing and evaluating our DL model contain CXRs for the following three categories: normal CXR (NORMAL), non-COVID-19 pneumonia CXR (PNEUMONIA), and COVID-19 pneumonia CXR (COVID). Our DL model was developed using two public (COVIDx and COVIDBIMCV) and one private (COVIDprivate) datasets. One public dataset (COVIDx) was built to accelerate the development of highly accurate and practical deep learning model for detecting COVID-19 cases (https://github.com/lindawangg/COVID-Net/blob/master/docs/COVIDx.md)15. The other public dataset (COVIDBIMCV) was constructed from two public datasets: the PadChest dataset (https://github.com/auriml/Rx-thorax-automatic-captioning)16 and BIMCV-COVID19+dataset (https://github.com/BIMCV-CSUSP/BIMCV-COVID-19)17. COVIDprivate was based on the dataset collected from six hospitals previously, and the two public datasets (COVIDx and COVIDBIMCV) were the same as those in previous studies18,19. The details of these datasets are described in the Supplementary material. Compared with the previous study, CXRs were added for COVIDprivate in the current study. The additional CXRs included 37, 7, and 31 cases of NORMAL, PNEUMONIA, and COVID, respectively. COVIDprivate contained 530 CXRs (176 NORMAL, 146 PNEUMONIA, and 208 COVID).

In addition to COVIDprivate, CXRs were collected from two other medical institutions. In total, 168 CXRs (80 NORMAL, 37 PNEUMONIA, and 51 COVID) collected from one medical institution (Hospital A) were used for the internal validation of the DL model (as a part of validation set) and for radiologists reading practice conducted before the observer study. Moreover, as unseen test set, 180 CXR cases (60 NORMAL, 60 PNEUMONIA, and 60 COVID) collected from another medical institution (Hospital B) were used for the external validation of the DL model and observer study of radiologists.

In the Hospital B, COVID was limited to those diagnosed with COVID-19 pneumonia using RT-PCR, and CXR was obtained after symptom onset. The time of COVID-19 diagnosis was between January 24, 2020, and May 5, 2020. PNEUMONIA was defined as patients clinically diagnosed with bacterial pneumonia that improved with appropriate treatment. Patients who showed no pneumonia on CT or had lung metastasis of malignancy and acute exacerbation of interstitial pneumonia were excluded from PNEUMONIA. NORMAL was defined as the absence of abnormalities in the lung, mediastinum, thoracic cavity, or chest wall on CXR and CT. NORMAL and PNEUMONIA were limited to cases before the summer of 2019 (before the COVID-19 pandemic). The details of the unseen test set collected from the Hospital B are described in the Supplementary material. The inclusion criteria of CXRs in the COVIDprivate and the Hospital A were the same as the previous study19.

Table 1 lists the details of each CXR dataset. The 180 cases (as the unseen test set) used for the external validation and reading sessions were adults aged 20years or older. In the 180 cases, NORMAL included 39 men and 21 women aged 58.127.9years. PNEUMONIA included 43 men and 17 women aged 76.220.8years. The COVID group included 46 men and 14 women aged 53.438.6years.

Our EfficientNet-based DL model was constructed in the same manner as described in previous papers18,19. Figure1 shows a schematic of the construction of the DL model. There are two major differences in the DL model construction between the present study and previous studies; one is that the 168 CXRs collected from Hospital A were used for internal validation as a part of the validation set, and the other is that the 180 CXRs collected from Hospital B were used for external validation as the unseen test set. The DL model development set included two public datasets, COVIDprivate, and 168 CXRs collected from Hospital A. Five different random divisions of the training and validation sets were created from the development set. In the division, 300, 300, and 90 images were randomly selected as the validation set from COVIDx, COVIDBIMCV, and COVIDprivate, respectively. The remaining images of COVIDx, COVIDBIMCV, and COVIDprivate were used as the train set. In addition, all the 168 CXRs collected from Hospital A were used for the validation set. Model training and internal validation of diagnostic performance were performed for the training set and validation set, respectively. The training of our DL model is also described in the Supplementary material.

Schematic illustration of dataset splitting and model training for our DL model. Abbreviation: DL, deep learning; COVIDx, public dataset used for COVID-Net; COVIDBIMCV, public dataset obtained from the PadChest and BIMCV-COVID19+datasets; COVIDprivate, private dataset collected from six hospitals; Hospital A, dataset collected for internal validation and radiologists practice before the observer study; Hospital B, dataset collected for external validation.

The inference results of the DL model were calculated using an ensemble of five trained models. For the 180 CXRs of the external validation, an average of the probabilities obtained from the five trained models was calculated as the inference results of the DL model to evaluate the diagnostic performance of the DL model and to provide supporting information for radiologists during the observer study.

The DL model calculated the probability of NORMAL, PNEUMONIA, or COVID for each CXR, with a total of 100%. We also created images using Grad-CAM and Grad-CAM++as explainable artificial intelligence, which visualized the reasoning for the diagnosis of the DL model20,21. Grad-CAM and Grad-CAM++images were used for the observer study. Minmax normalization with a linear transformation was performed on the original Grad-CAM and Grad-CAM++images.

Eight radiologists (with 520years of experience in diagnostic radiology) performed the observer study at two medical facilities. For the 180 CXRs collected from Hospital B, each radiologist performed two reading sessions over a period of more than 1month. One reading session was performed with reference to CXRs only, and the other was performed with reference to both CXRs and the results of the DL model. The order of the two sessions was randomly selected to reduce bias. The eight radiologists scored the probabilities of NORMAL, PNEUMONIA, and COVID on a 100% scale. In the reading session with the DL model, the radiologists referred to the probabilities of NORMAL, PNEUMONIA, and COVID calculated using the DL model. If there was any uncertainty regarding the probabilities of the DL model, the results of Grad-CAM and Grad-CAM++were available. Images of the 168 CXRs collected from Hospital A were also processed with Grad-CAM and Grad-CAM++, and the diagnosis of the DL model and images of Grad-CAM and Grad-CAM++of the 168 CXRs were presented to the radiologists for practice sessions before each reading session. Eight radiologists were taught how to interpret the Grad-CAM and Grad-CAM++images before the observer study. There was no time limit for reading and practice sessions. Prior to the reading sessions, only the approximate frequencies of the three categories were presented to the radiologists and no other clinical information was provided. Our novelties in this study were to investigate whether radiologists changed their diagnosis by referring to our DL model of CXR and whether the diagnostic performance of radiologists was significantly improved.

After the observer study, one senior radiologist visually evaluated the 180 Grad-CAM++images in the test set. The visual evaluation of the Grad-CAM++images was performed on the images that were accurately diagnosed by the DL. The radiologist visually examined the CXR and Grad-CAM++images and determined whether the Grad-CAM++images were typical or understandable. The typical Grad-CAM++images were described in Supplementary material. If abnormal findings on CXR images were highlighted on Grad-CAM++images, the cases were considered understandable by the radiologist. In addition, for COVID, the radiologist counted the number of Grad-CAM++images with highlighted regions outside the lung area.

We evaluated the diagnostic performance of the DL model alone and compared the results between reading sessions with and without the DL model. The evaluation metrics were accuracy, sensitivity, specificity, and area under the curve (AUC) in the receiver operating characteristics. Because three-category classification was performed, these metrics were calculated class-wise (one-vs-rest), except for accuracy. For the AUC, multi-reader multi-case statistical analysis was used to statistically analyze the results of the eight radiologists. MRMCaov was used for the statistical analyses22. Although MRMCaov is a statistical method designed for binary classification of two categories, this study was designed to diagnose three categories: NORMAL, PNEUMONIA, and COVID. Therefore, the three-category classification was divided into three binary classifications (one-vs-rest): (1) NORMAL versus PNEUMONIA or COVID, (2) PNEUMONIA versus NORMAL or COVID, and (3) COVID versus NORMAL or PNEUMONIA. We then compared the class-wise AUC of the eight radiologists between reading sessions with and without the DL model. The difference in the AUC was statistically tested using MRMCaov. Because it was necessary to integrate the results from the eight radiologists, the class-wise MRMCaov was used in the present study. To control the family-wise error rate, Bonferroni correction was used; a p value less than 0.01666 was considered statistically significant. R (version 4.1.2) was used for the statistical analysis.

See the article here:

Computer-aided diagnosis of chest X-ray for COVID-19 diagnosis in ... - Nature.com