lOMoARcPSD|44249588

UNIT 5 - FUNDAMENTALS OF HEALTHCARE ANALYTICS

REGULATION 2021
 lOMoARcPSD|44249588

UNIT V CASE STUDIES

Epidemical reading and interpreting of epidemical studies

The healthcare industry has increasingly turned to data analytics to address epidemiological
challenges, especially in the wake of the COVID-19 pandemic. By analyzing large volumes of
data, healthcare professionals can identify trends, predict outbreaks, optimize resource allocation,
and ultimately improve patient outcomes. Let's explore the detailed case studies that showcase
how healthcare analytics is applied in epidemiology.

1. Big Data in Healthcare Analytics

Big data analytics involves analyzing vast datasets to derive meaningful insights. In healthcare,
big data comes from various sources, including hospital records, patient medical records, clinical
trials, and more. A key benefit of this approach is the ability to analyze data in real-time,
allowing healthcare professionals to respond swiftly to emerging trends.
A specific case study demonstrates the potential of big data in healthcare. It involves analyzing
electronic health records (EHR) to identify trends in patient care. This analysis helped to improve
patient outcomes by optimizing treatment plans and reducing unnecessary procedures.

2. Real-Time Health Monitoring Systems

Real-time data analysis is crucial in health monitoring systems. These systems continuously
collect data from wearable devices, sensors, and other medical equipment, enabling healthcare
professionals to monitor patients' health in real time.
A case study focusing on real-time health monitoring revealed how data analytics can be used to
predict critical events. By analyzing heart rate, blood pressure, and other vital signs, medical
professionals were able to predict heart attacks and other emergencies, enabling early
intervention and potentially saving lives.

3. Fighting Pandemics with Big Data Analytics

The COVID-19 pandemic underscored the importance of big data analytics in healthcare. A case
study exploring the role of big data during the pandemic highlighted how data-driven insights
helped track the spread of the virus, predict outbreaks, and develop vaccines.
 lOMoARcPSD|44249588

In another case study, big data analytics played a significant role in contact tracing, allowing
public health officials to identify and isolate infected individuals quickly. This approach
significantly reduced the spread of COVID-19 and helped manage the pandemic.

4. Using Machine Learning in Healthcare Analytics

Machine learning (ML) is a subset of artificial intelligence that allows computers to learn from
data without explicit programming. A case study on a healthcare platform using big data
analytics and machine learning demonstrated how ML algorithms could analyze patient data to
predict disease progression and recommend personalized treatment plans.
In another case study, ML algorithms were used to analyze medical imaging data to detect early
signs of diseases like cancer. This approach enabled early diagnosis and treatment, significantly
improving patient outcomes.

5. Real-Life Examples of Big Data in Healthcare

Big data in healthcare provides numerous real-life applications. A detailed case study explored
various use cases, such as predictive analytics for patient readmissions, analysis of large clinical
datasets to identify trends, and personalized medicine approaches based on genetic data.
One specific example involves analyzing hospital data to predict patient readmissions. By
examining factors such as age, underlying health conditions, and prior hospitalizations,
healthcare professionals can identify patients at high risk of readmission and implement
preventive measures.

Application in community health

Community health refers to simple health services that are delivered by laymen outside hospitals
and clinics. Community health is also the subset of public health that is taught to and practiced
by clinicians as part of their normal duties. Community health volunteers and community health
workers work with primary care providers to facilitate entry into, exit from and utilization of the
formal health system by community members.

Community health is a major field of study within the medical and clinical sciences which
focuses on the maintenance, protection, and improvement of the health status of population
groups and communities. It is a distinct field of study that may be taught within a separate school
of public health or Preventive Healthcare. The WHO defines community health as:

Downloaded by Senthil Kumar K (maksen79@gmail.com)

 lOMoARcPSD|44249588

Environmental, Social, and Economic resources to sustain emotional and physical well being
among people in ways that advance their aspirations and satisfy their needs in their unique
environment.

Medical interventions that occur in communities can be classified as three categories: Primary
care, Secondary care, and Tertiary care. Each category focuses on a different level and approach
towards the community or population group. In the United States, Community health is rooted
within Primary healthcare achievements. Primary healthcare programs aim to reduce risk factors
and increase health promotion and prevention. Secondary healthcare is related to "hospital care"
where acute care is administered in a hospital department setting. Tertiary healthcare refers to
highly specialized care usually involving disease or disability management.

A community health prayer by Dr Edmond Fernandes

The success of community health programs relies upon the transfer of information from health
professionals to the general public using one-to-one or one-to-many communication (mass
communication). The latest shift is towards health marketing.

Community health services are classified into categories including:

Preventive health services such as chemoprophylaxis for Tuberculosis, cancer screening and
treatment of diabetes and hypertension.
Promotive health services such as Health education, family planning, vaccination and nutritional
supplementation
Curative health services such as treatment of jiggers, lice infestation, Malaria and Pneumonia.
Rehabilitative health services such as provision of prosthetics, Social work, Occupational
therapy, Physical therapy, Counselling and other Mental health services.
Community health workers and volunteers
Community health workers (also known as community health assistants and community health
officers) are local public health workers with a deep understanding of their community's health
needs and challenges. They serve as a bridge between their community and local health systems
to ensure high quality and culturally competent service delivery. They have vocational,
professional or academic qualifications which enable them to provide training, supervisory,
administrative, teaching and research services in community health departments.
 lOMoARcPSD|44249588

Community health volunteers are members of a local community who have experience and
training on the health problems prevalent in their community and care services available, in order
to identify and link those in need with local providers. Community health volunteers may be
referred to by different titles depending on their local health system; these titles can included lay
health workers, health volunteers, village health agents, non-specialist healthcare providers, and
village health agents.

Community health volunteers provide basic services such as distribution of water chlorination
tablets, mosquito nets and health education material. They will involve or work with registered
clinicians when they encounter sick or recovering patients or those with complex or ongoing
needs.

Community health organizations are non-profit and non government organization which
administers and coordinates the delivery of health care services to people living in a designated
community or neighborhood. It helps people understand their status of health or social
conditions. Providing advocacy for those who need it and holding groups and individual
meetings with people in the community. The vital role is advocating for the rights and interests of
their community members. They raise awareness about issues affecting their community by
research, dialogues and lobby for policies and programs that address those issues.

Measuring community health

Community health is generally measured by Geographical Information Systems and
Demographic data. Geographic Information Systems can be used to define sub-communities
when neighborhood location data is not enough. Traditionally Community health has been
measured using sampling data which was then compared to well-known data sets, like the
National Health Interview Survey or National Health and Nutrition Examination Survey. With
technological development, information systems could store more data for small-scale
communities, cities, and towns; as opposed to census data that only generalize information about
small populations based on the overall population. Geographical Information Systems (GIS) can
give more precise information about community resources, even at neighborhood levels. The
ease of use of Geographic Information Systems (GIS), advances in multilevel statistics, and
spatial analysis methods make it easier for researchers to procure and generate data related to the
built environment.

Social media can also play a big role in health information analytics. Studies have found social
media being capable of influencing people to change their unhealthy behaviors and encourage
interventions capable of improving health status. Social media statistics combined with
 lOMoARcPSD|44249588

Geographical Information Systems (GIS) may provide researchers with a more complete image
of community standards for health and well being.

Categories of community health

Primary Healthcare and Primary Prevention
Community-based health promotion emphasizes Primary Prevention and population-based
perspective (traditional prevention). It is the goal of Community Health to have individuals in a
certain community improve their lifestyle or seek medical attention. Primary Healthcare is
provided by health professionals, specifically the ones a patient sees first that may refer them to
Secondary or Tertiary care.

Primary prevention refers to the early avoidance and identification of risk factors that may lead
to certain diseases and disabilities. Community-focused efforts including immunizations,
classroom teaching, and awareness campaigns are all good examples of how primary prevention
techniques are utilized by communities to change certain health behaviors. Prevention programs,
if carefully designed and drafted, can effectively prevent problems that children and adolescents
face as they grow up. This finding also applies to all groups and classes of people. Prevention
programs are one of the most effective tools health professionals can use to significantly impact
individual, population, and community health.

Secondary Healthcare and Secondary Prevention

Community health can also be improved with improvements in individuals' environments.
Community health status is determined by the environmental characteristics, behavioral
characteristics, social cohesion in the environment of that community. Appropriate modifications
in the environment can help to prevent unhealthy behaviors and negative health outcomes.

Secondary prevention refers to improvements made in a patient's lifestyle or environment after

the onset of disease or disability. This sort of prevention works to make life easier for the patient
since it is too late to prevent them from their current disease or disability. An example of
secondary prevention is when those with occupational low back pain are provided with strategies
to stop their health status from worsening; the prospects of secondary prevention may even hold
more promise than primary prevention in this case.[16]

Tertiary Healthcare
In Tertiary healthcare, community health can only be affected with professional medical care
involving the entire population. Patients need to be referred to specialists and undergo advanced
 lOMoARcPSD|44249588

medical treatment. In some countries, there are more sub-specialties of medical professions than
there are primary care specialists. Health inequalities are directly related to social advantage and
social resources.

s
The Evolution of Medical Imaging and Growing Data
The medical imaging field has evolved tremendously over the past few decades. As technologies
like CT, MRI, and PET scans were introduced and adopted, the amount of data produced
skyrocketed. A single CT scan can generate over 1,000 images, and the average large hospital
now produces over 50 petabytes of data each year.

The Rise of AI and Analytics

To make sense of this massive amount of information, healthcare providers have turned to
artificial intelligence and analytics. Machine learning algorithms can detect patterns and insights
that humans may miss. Analytics tools are helping radiologists diagnose conditions more
accurately and detect diseases earlier. For example, AI systems can analyze CT scans to diagnose
a stroke, detect signs of Alzheimer's in brain scans years before symptoms appear, or spot lung
cancers in X-rays with greater accuracy than human radiologists.

AI and analytics are also making medical imaging more efficient. Systems can automatically
label, sort, and prioritize scans to reduce the time radiologists spend on routine tasks.
Radiologists then have more time to focus on complex cases. AI tools can also suggest possible
diagnoses or areas of concern on a scan to help guide the radiologist.
 lOMoARcPSD|44249588

While AI and analytics will not replace radiologists, they are transforming medical imaging.
Radiologists working with these advanced tools have access to insights that can help improve
care, reduce errors, and change medical outcomes for the better. The future of diagnostic
imaging is data-driven, and AI and analytics are helping unlock the potential of all that data to
benefit patients and physicians.

How Analytics Is Transforming Medical Imaging

These days, medical imaging is so much more than just capturing pictures of the human body.
Thanks to advanced analytics, radiologists and physicians can now gain the following insights to
help improve patient diagnosis and care.:

Detecting Disease Earlier

Analytics allows radiologists to compare scans to detect even the smallest changes over time.
This helps spot disease progression much earlier, even before symptoms appear. For example, by
comparing brain scans from year to year, doctors can detect signs of Alzheimer's disease or other
dementias far sooner.

Personalized Treatment Plans

By analyzing a patient's scans and medical history along with those of similar patients, doctors
can develop customized treatment plans tailored to their needs. For instance, analyzing lung CT
scans from thousands of people with similar cancers helps oncologists determine the therapy
most likely to benefit a specific patient based on subtle differences in their tumors.

AI-Assisted Diagnosis
Artificial intelligence and machine learning tools help radiologists analyze scans faster and more
accurately. AI systems can detect patterns that humans might miss, helping diagnose conditions
like pneumonia, stroke, or fractures. Some tools even suggest possible diagnoses and highlight
areas of concern on the scans. While AI won't replace radiologists, it makes them much more
efficient and helps reduce errors.

Continuous Monitoring
New wearable and implantable sensors generate ongoing data that, when combined with
analytics, enable continuous patient monitoring without frequent office visits or scans. For
example, sensors can track vital signs, activity levels, and other metrics in people with chronic
heart failure, Alzheimer's, or Parkinson's and alert doctors to changes that may require
intervention.

Analytics is transforming medical imaging into a powerful tool for precision diagnosis,
treatment, and monitoring. By unlocking insights from imaging data, analytics helps doctors
provide the right care at the right time for every patient. The future of healthcare will rely on
these kinds of intelligent, data-driven approaches.

Key Areas Where Analytics Is Improving Imaging Workflows

Analytics is revolutionizing imaging in healthcare by turning raw data into actionable insights.
Three key areas seeing major improvements are:
 lOMoARcPSD|44249588

Workflow optimization
By analyzing historical imaging data, systems can predict how resources are utilized and find
ways to streamline the process. For example, AI may determine the optimal scheduling approach
to minimize wait times and maximize scanner usage. It can also automatically pre-fetch patient
data and reports to save staff time.

Enhanced diagnostics
Computer-aided detection (CAD) tools analyze scans to identify possible abnormalities, like
tumors or lesions, that radiologists may have missed. CAD acts as a second set of eyes to ensure
the most accurate diagnosis. It is particularly useful for complex or subtle cases.

Personalized care
In the future, AI and analytics may tap into a patient's full medical history, genetics, lifestyle
habits, and more to develop customized diagnosis and treatment plans tailored to their unique
needs. For example, certain biomarkers or characteristics identified in a patient's MRI or CT scan
could indicate a higher risk of disease or better response to specific therapies. Physicians can
then make precision medicine decisions based on data-driven insights.

Some Real-world Case studies

Case studies show how analytics applied to medical imaging data is transforming diagnosis and
treatment:

Stanford University: Lung Cancer Detection

A study from Stanford University analyzed CT scans from patients at high risk of lung cancer
using a deep learning algorithm. The AI system detected lung nodules that radiologists had
missed, identifying lung cancers at an early, more treatable stage. Early detection of lung cancer
can improve 5-year survival rates from under 20% to over 90%.

Harvard and MIT: Tailored Treatment Plans

Researchers at Harvard and MIT developed a machine learning model analyzing MRI scans that
can predict how individuals will respond to certain cancer treatments. The model analyzes tumor
shape, texture, and location to determine who will benefit from chemotherapy before starting
treatment. This “precision oncology” approach helps doctors create tailored treatment plans for
each patient.

University of California: Eliminating False Positives in Medical Imaging

False positives in medical imaging often led to unnecessary follow-up tests and procedures that
are costly and stressful for patients. A study from the University of California, San Francisco
used a deep learning model to analyze mammograms. The AI system reduced false positive rates
by up to 85% while still detecting breast cancers with high accuracy. Decreasing false positives
in mammography and other screening tools may make these tests more efficient and less anxiety-
provoking for patients.

In these examples and others, machine learning and data analytics are making medical imaging
smarter, faster, and more accurate. As the field progresses, AI promises to boost the early
 lOMoARcPSD|44249588

detection of disease, personalize diagnosis and treatment, reduce errors, lower costs and
ultimately improve patient outcomes.

The future of healthcare will be driven not just by new technologies but by innovative use of the
massive amounts of data already at our fingertips. Analytics is transforming that data into life-
saving knowledge.

The Importance of Data Security and Privacy with Healthcare Analytics

When it comes to healthcare data and analytics, the utmost priority lies in ensuring privacy and
security. Patients entrust their healthcare providers with intimate details about their health and
lives, expecting a commitment to the confidentiality of this information. Following robust
measures must be in place to safeguard privacy:

Protecting Patient Data

Healthcare organizations have a legal and ethical obligation to keep patient data private and
secure. Strict data governance policies and security measures must be put in place to protect
sensitive information from unauthorized access or misuse. Strong encryption, multi-factor
authentication, and restricting data access to only those who truly need it are essential
safeguards.

Regular audits and risk assessments help identify and address any vulnerabilities. If a data breach
were to occur, healthcare providers must have response plans ready to act quickly to contain the
incident, determine what information was compromised, and notify any patients that may have
been affected.

Building Trust
For patients to feel comfortable sharing their health data, they need to have confidence in how
it's handled and used. Transparency about data collection, storage, access, and analytics policies
helps build that trust. Patients should understand who sees their data, and for what purposes, and
be assured it will only be used to benefit their care. They must also maintain some control over
their data, with options to opt out of certain data uses if desired.

When data is used responsibly and for the right reasons, analytics can transform medical care and
improve patient outcomes. But with sensitive health information, privacy and security will
always need to come first.

Healthcare organizations must earn and maintain the trust of their patients through
accountability, transparency, and a steadfast commitment to safeguarding data. The benefits of
healthcare analytics depend on it.
 lOMoARcPSD|44249588

Case study on respiratory data

Case Study: 33-Year-Old Female Presents with Chronic SOB and Cough

Case Presentation
History of Present Illness: A 33-year-old white female presents after admission to the general
medical/surgical hospital ward with a chief complaint of shortness of breath on exertion. She
reports that she was seen for similar symptoms previously at her primary care physician’s office
six months ago. At that time, she was diagnosed with acute bronchitis and treated with
bronchodilators, empiric antibiotics, and a short course oral steroid taper. This management did
not improve her symptoms, and she has gradually worsened over six months. She reports a 20-
pound (9 kg) intentional weight loss over the past year. She denies camping, spelunking, or
hunting activities. She denies any sick contacts. A brief review of systems is negative for fever,
night sweats, palpitations, chest pain, nausea, vomiting, diarrhea, constipation, abdominal pain,
neural sensation changes, muscular changes, and increased bruising or bleeding. She admits a
cough, shortness of breath, and shortness of breath on exertion.

Social History: Her tobacco use is 33 pack-years; however, she quit smoking shortly prior to the
onset of symptoms, six months ago. She denies alcohol and illicit drug use. She is in a married,
monogamous relationship and has three children aged 15 months to 5 years. She is employed in a
cookie bakery. She has two pet doves. She traveled to Mexico for a one-week vacation one year
ago.

Allergies: No known medicine, food, or environmental allergies.

Past Medical History: Hypertension
Past Surgical History: Cholecystectomy
Medications: Lisinopril 10 mg by mouth every day
Physical Exam:
Vitals: Temperature, 97.8 F; heart rate 88; respiratory rate, 22; blood pressure 130/86; body mass
index, 28
General: She is well appearing but anxious, a pleasant female lying on a hospital stretcher. She is
conversing freely, with respiratory distress causing her to stop mid-sentence.
Respiratory: She has diffuse rales and mild wheezing; tachypneic.
Cardiovascular: She has a regular rate and rhythm with no murmurs, rubs, or gallops.
Gastrointestinal: Bowel sounds X4. No bruits or pulsatile mass.

Initial Evaluation
Laboratory Studies: Initial work-up from the emergency department revealed pancytopenia with
a platelet count of 74,000 per mm3; hemoglobin, 8.3 g per and mild transaminase elevation, AST
90 and ALT 112. Blood cultures were drawn and currently negative for bacterial growth or Gram
staining.

Chest X-ray
 lOMoARcPSD|44249588

Impression: Mild interstitial pneumonitis

Differential Diagnosis
Aspiration pneumonitis and pneumonia
Bacterial pneumonia
Immunodeficiency state and Pneumocystis jiroveci pneumonia
Carcinoid lung tumors
Tuberculosis
Viral pneumonia
Chlamydial pneumonia
Coccidioidomycosis and valley fever
Recurrent Legionella pneumonia
Mediastinal cysts
Mediastinal lymphoma
Recurrent mycoplasma infection
Pancoast syndrome
Pneumococcal infection
Sarcoidosis
Small cell lung cancer
Aspergillosis
Blastomycosis
Histoplasmosis
Actinomycosis

Confirmatory Evaluation
CT of the chest was performed to further the pulmonary diagnosis; it showed a diffuse
centrilobular micronodular pattern without focal consolidation.

On finding pulmonary consolidation on the CT of the chest, a pulmonary consultation was

obtained. Further history was taken, which revealed that she has two pet doves. As this was her
third day of broad-spectrum antibiotics for a bacterial infection and she was not getting better, it
was decided to perform diagnostic bronchoscopy of the lungs with bronchoalveolar lavage to
look for any atypical or rare infections and to rule out malignancy (Image 1).

Bronchoalveolar lavage returned with a fluid that was cloudy and muddy in appearance. There
was no bleeding. Cytology showed Histoplasma capsulatum.

Diagnosis
Based on the bronchoscopic findings, a diagnosis of acute pulmonary histoplasmosis in an
immunocompetent patient was made.
 lOMoARcPSD|44249588

Management
Pulmonary histoplasmosis in asymptomatic patients is self-resolving and requires no treatment.
However, once symptoms develop, such as in our above patient, a decision to treat needs to be
made. In mild, tolerable cases, no treatment other than close monitoring is necessary. However,
once symptoms progress to moderate or severe, or if they are prolonged for greater than four
weeks, treatment with itraconazole is indicated. The anticipated duration is 6 to 12 weeks total.
The response should be monitored with a chest x-ray. Furthermore, observation for recurrence is
necessary for several years following the diagnosis. If the illness is determined to be severe or
does not respond to itraconazole, amphotericin B should be initiated for a minimum of 2 weeks,
but up to 1 year. Cotreatment with methylprednisolone is indicated to improve pulmonary
compliance and reduce inflammation, thus improving work of respiration.

Discussion
Histoplasmosis, also known as Darling disease, Ohio valley disease, reticuloendotheliosis,
caver's disease, and spelunker's lung, is a disease caused by the dimorphic fungi Histoplasma
capsulatum native to the Ohio, Missouri, and Mississippi River valleys of the United States. The
two phases of Histoplasma are the mycelial phase and the yeast phase.

Etiology/Pathophysiology

Histoplasmosis is caused by inhaling the microconidia of Histoplasma spp. fungus into the lungs.
The mycelial phase is present at ambient temperature in the environment, and upon exposure to
37 C, such as in a host’s lungs, it changes into budding yeast cells. This transition is an important
determinant in the establishment of infection. Inhalation from soil is a major route of
transmission leading to infection. Human-to-human transmission has not been reported. Infected
individuals may harbor many yeast-forming colonies chronically, which remain viable for years
after initial inoculation. The finding that individuals who have moved or traveled from endemic
to non-endemic areas may exhibit a reactivated infection after many months to years supports
this long-term viability. However, the precise mechanism of reactivation in chronic carriers
remains unknown.

Infection ranges from an asymptomatic illness to a life-threatening disease, depending on the

host’s immunological status, fungal inoculum size, and other factors. Histoplasma spp. have
grown particularly well in organic matter enriched with bird or bat excrement, leading to the
association that spelunking in bat-feces-rich caves increases the risk of infection. Likewise,
ownership of pet birds increases the rate of inoculation. In our case, the patient did travel outside
of Nebraska within the last year and owned two birds; these are her primary increased risk
factors.
 lOMoARcPSD|44249588

Non-immunocompromised patients present with a self-limited respiratory infection. However,

the infection in immunocompromised hosts disseminated histoplasmosis progresses very
aggressively. Within a few days, histoplasmosis can reach a fatality rate of 100% if not treated
aggressively and appropriately. Pulmonary histoplasmosis may progress to a systemic infection.
Like its pulmonary counterpart, the disseminated infection is related to exposure to soil
containing infectious yeast. The disseminated disease progresses more slowly in
immunocompetent hosts compared to immunocompromised hosts. However, if the infection is
not treated, fatality rates are similar. The pathophysiology for disseminated disease is that once
inhaled, Histoplasma yeast are ingested by macrophages. The macrophages travel into the
lymphatic system where the disease, if not contained, spreads to different organs in a linear
fashion following the lymphatic system and ultimately into the systemic circulation. Once this
occurs, a full spectrum of disease is possible. Inside the macrophage, this fungus is contained in
a phagosome. It requires thiamine for continued development and growth and will consume
systemic thiamine. In immunocompetent hosts, strong cellular immunity, including
macrophages, epithelial, and lymphocytes, surround the yeast buds to keep infection localized.
Eventually, it will become calcified as granulomatous tissue. In immunocompromised hosts, the
organisms disseminate to the reticuloendothelial system, leading to progressive disseminated
histoplasmosis.

Symptoms of infection typically begin to show within three to17 days. Immunocompetent
individuals often have clinically silent manifestations with no apparent ill effects. The acute
phase of infection presents as nonspecific respiratory symptoms, including cough and flu. A
chest x-ray is read as normal in 40% to 70% of cases. Chronic infection can resemble
tuberculosis with granulomatous changes or cavitation. The disseminated illness can lead to
hepatosplenomegaly, adrenal enlargement, and lymphadenopathy. The infected sites usually
calcify as they heal. Histoplasmosis is one of the most common causes of mediastinitis.
Presentation of the disease may vary as any other organ in the body may be affected by the
disseminated infection.

Diagnosis

The clinical presentation of the disease has a wide-spectrum presentation which makes diagnosis
difficult. The mild pulmonary illness may appear as a flu-like illness. The severe form includes
chronic pulmonary manifestation, which may occur in the presence of underlying lung disease.
The disseminated form is characterized by the spread of the organism to extrapulmonary sites
with proportional findings on imaging or laboratory studies. The Gold standard for establishing
the diagnosis of histoplasmosis is through culturing the organism. However, diagnosis can be
established by histological analysis of samples containing the organism taken from infected
organs. It can be diagnosed by antigen detection in blood or urine, PCR, or enzyme-linked
immunosorbent assay. The diagnosis also can be made by testing for antibodies again the fungus.

Treatment
 lOMoARcPSD|44249588

Pulmonary histoplasmosis in asymptomatic patients is self-resolving and requires no treatment.

However, once symptoms develop, such as in our above patient, a decision to treat needs to be
made. In mild, tolerable cases, no treatment other than close monitoring is necessary. However,
once symptoms progress to moderate or severe or if they are prolonged for greater than four
weeks, treatment with itraconazole is indicated. The anticipated duration is 6 to 12 weeks. The
patient's response should be monitored with a chest x-ray. Furthermore, observation for
recurrence is necessary for several years following the diagnosis. If the illness is determined to
be severe or does not respond to itraconazole, amphotericin B should be initiated for a minimum
of 2 weeks, but up to 1 year. Cotreatment with methylprednisolone is indicated to improve
pulmonary compliance and reduce inflammation, thus improving the work of respiration.

The disseminated disease requires similar systemic antifungal therapy to pulmonary infection.
Additionally, procedural intervention may be necessary, depending on the site of dissemination,
to include thoracentesis, pericardiocentesis, or abdominocentesis. Ocular involvement requires
steroid treatment additions and necessitates ophthalmology consultation. In pericarditis patients,
antifungals are contraindicated because the subsequent inflammatory reaction from therapy
would worsen pericarditis.

Patients may necessitate intensive care unit placement dependent on their respiratory status, as
they may pose a risk for rapid decompensation. Should this occur, respiratory support is
necessary, including non-invasive BiPAP or invasive mechanical intubation. Surgical
interventions are rarely warranted; however, bronchoscopy is useful as both a diagnostic measure
to collect sputum samples from the lung and therapeutic to clear excess secretions from the
alveoli. Patients are at risk for developing a coexistent bacterial infection, and appropriate
antibiotics should be considered after 2 to 4 months of known infection if symptoms are still
present.

Prognosis

If not treated appropriately and in a timely fashion, the disease can be fatal, and complications
will arise, such as recurrent pneumonia leading to respiratory failure, superior vena cava
syndrome, fibrosing mediastinitis, pulmonary vessel obstruction leading to pulmonary
hypertension and right-sided heart failure, and progressive fibrosis of lymph nodes. Acute
pulmonary histoplasmosis usually has a good outcome on symptomatic therapy alone, with 90%
of patients being asymptomatic. Disseminated histoplasmosis, if untreated, results in death
within 2 to 24 months. Overall, there is a relapse rate of 50% in acute disseminated
histoplasmosis. In chronic treatment, however, this relapse rate decreases to 10% to 20%. Death
is imminent without treatment.

Case 1: Mild Asthma Symtpoms

Q: TR, a 23-year-old man, is seeking advice. Although he is generally in good health, he has
recently been developing a mild wheezing and tightness in his chest after exercising and playing
 lOMoARcPSD|44249588

soccer. TR does not smoke and has not had any respiratory illnesses recently. However, he does
recall using an inhaler as a child for asthma but was told that he had outgrown the condition. TR
does not take any medications. He says he does not want to follow up with a physician, if
possible, because he is between jobs and does not have health care insurance. What information
can you provide regarding nonpharmacologic approaches to managing this condition and self-
care?

A: TR may be suffering from intermittent, exercise-induced asthma symptoms, given his history
and symptoms. As far as self-care options, it is important to recognize that the FDA recently
approved a new formulation of the OTC epinephrine inhaler Primatene Mist to treat mild,
intermittent asthma symptoms. This chlorofluorocarbon-, propellant-free formulation is likely to
be available in early 2019 on pharmacy shelves and will replace the original formulation
withdrawn from the market in 2011. Approved for use in patients 12 years and older, unlike the
original preparation, the new formulation will include a different inhaler device accompanying
the active ingredient, epinephrine. Patients should be educated about proper use, including the
need for priming and shaking before use and cleaning the device afterward. Although this is a
reasonable option for alleviating TR’s infrequent symptoms, it is nonetheless worth educating
him about the chronic and inflammatory nature of asthma. If TR does not respond within 20
minutes of using a dose, experiences more than 2 attacks in a week, notices worsening
symptoms, or is using more than 8 doses in a 24-hour period, he should be urged to seek medical
evaluation. This medication is an effective bronchodilator but will ultimately not address more
moderate-to-severe asthma symptoms and treat the underlying disease.1

Case 2: Cold Symptoms

Q: AI, a 60-year-old woman, wants to know what OTC remedy is best for treating her cold
symptoms. Her medical history includes hypertension and hyperthyroidism, and she is taking
several medications, including lisinopril and methimazole. AI is suffering from significant nasal
congestion, which is preventing her from sleeping well at night, and she would like to take a
nonprescription decongestant, but the label on the medication she selected says to check with a
doctor or a pharmacist if individuals have high blood pressure or thyroid disease. What
recommendations can you provide?

A: There are an estimated 62 million occurrences of the common cold in the United States each
year. Symptoms include congestion, cough, headache, malaise, pain, postnasal drip, rhinorrhea,
sinus pressure, and/or sneezing. Given AI’s medical and medication history, she should avoid
systemic decongestants containing pseudoephedrine or other active ingredients that cause
vasoconstriction to provide symptom relief. These agents can cause increases in blood pressure
and may exacerbate symptoms of hyperthyroidism. To err on the side of caution, it is prudent to
recommend forms of nonpharmacologic relief of congestion in this case, including nasal
decongestant strips, saline or a teapot, or topical camphor- or menthol-containing preparations
that may help clear inflamed nasal passages.
 lOMoARcPSD|44249588

Case 3: Pearls for Patients with COPD

Q: HW, a 74-year-old man, is looking for advice. His wife encouraged him to speak with a
pharmacist, as he recently received a diagnosis of chronic obstructive pulmonary disease
(COPD) and wants to make sure he is doing everything he can to reduce his chances of
complicating or exacerbating his condition. HW has a significant medical history, including
coronary artery disease post stent placement, heart failure, hypertension, and high cholesterol,
along with his new diagnosis of COPD. He has an extensive medication list, which includes
various pharmacologic agents for managing these conditions. HW started smoking as a teenager
and quit cold turkey several years ago at the uArging of his cardiologist. What information can
you provide regarding self-management of HW’s COPD?

A: COPD is estimated to affect more than 16 million Americans, with older adults more likely to
experience this condition than younger individuals.2 Although COPD necessitates medical
evaluation and prescription drug treatment, self-care and self-management are critical to ensuring
optimization of symptom control and quality of life through reduction in exacerbations and
hospitalizations for complications. In HW’s case, there are several important educational
opportunities for the pharmacist. These include asking him to demonstrate his inhaler technique
and reviewing it with him; ensuring that he is up-to-date on all indicated vaccines, based on his
age and medical comorbidities; inquiring about medical follow-up appointments; and
underscoring the importance of using his maintenance medications, even if he is not having
trouble breathing. In addition, social support is critical to ensuring optimal health outcomes,
particularly with a chronic lung disease such as COPD. Living with a chronic disease is difficult,
and the risk of comorbid depression is high. Encourage HW to include his wife, other family
members, or caregivers to help him manage his condition and troubleshoot with him if questions
arise.

Case 4: InfluenzaQ: AT, a 38-year-old woman, calls the pharmacy inquiring about self-care for
influenzalike symptoms. She works as an elementary school teacher and reports a sudden onset
of chills, fatigue, general malaise, and muscle pains that caused her to be sent home out of fear
that she had the flu, even though she had received her influenza vaccine. Upon arriving home,
AT took her temperature and confirmed a fever of 101.6°F. She reports no significant medical
history and says she takes no chronic medications other than occasional nonprescription
medicines for mild conditions. AT has a cadre of cold and cough preparations at home, along
with some pain relievers, but she wants to know what will best alleviate her symptoms and help
her recover as quickly as possible. What recommendations or eAducation on self-care for
managing flulike symptoms can you offer?

A: Symptoms associated with the influenza virus can vary and range from mild to severe. Given
AT’s age and medical history, she is not considered high risk for developing influenza-associated
complications. Nonetheless, it isimportant to educate her about proper self-care to promote
recovery and reduce her chances of spreading the illness. First and foremost, encourage AT to
stay home and avoid contact with others as much as possible, unless she needs medical care. She
 lOMoARcPSD|44249588

may consider seeking medical evaluation for confirmation of influenza and treatment with
prescription medications to reduce symptom duration. For the 2018-2019 influenza season in the
United States, 4 antiviral medications (oral baloxavir and oseltamivir, inhaled zanamivir, and
intravenous peramivir) are approved and recommended to treat influenza.3 Regarding the
efficacy of these medications, time is of the essence, as these agents should be initiated within 48
hours of the onset of symptoms. Beyond consideration for antiviral therapy, supportive care
interventions should be used. These include controlling the fever with whatever analgesic or
antipyretic AT has in her medicine cabinet; hydrating with water, soups, juices, or other
noncaffeinated beverages; and washing her hands to avoid spreading the virus. Encourage her to
stay home for at least 24 hours after the fever subsides, in accordance with CDC
recommendations.

Case study on ECG data.

ECG dataset description

This retrospective study was approved by the Institutional Review Board (IRB) of Konkuk
University Medical Center on March 12, 2020 (KUH1010966). Informed consent was waived by
the IRB because the data were de-identified. This retrospective study was also approved by IRB
of Seoul National University Bundang Hospital on October 10, 2018(B-1808/484-001), March 7,
2019(B-1903/531-101). Informed consent was waived by the IRB because the data were de-
identified. The data from Konkuk University Medical Center was analyzed by researchers at
Seoul National University Bundang Hospital. Data from Seoul National University Bundang
Hospital were used for verification to derive the results of this paper. Finally, the logic of the
paper was developed using only the data from Konkuk University Medical Center.

We constructed a large standard 12-lead ECG dataset that included the data from 13,241 12-lead
ECG tests that were performed at Konkuk University Hospital in South Korea.
The data were categorized according to the following 14 diagnoses:
-Sinus rhythm (normal)
-Atrial fibrillation (AF)
-Atrial flutter (AFL)
-Atrioventricular block (AV block)
-Supraventricular tachycardia (SVT)
-Ventricular tachycardia (VT)
-Ventricular premature contraction (VPC)
-Atrial premature contraction (APC)
-Sinus pause (SP)
-Marked sinus bradycardia
-Atrial tachycardia
-Junctional premature beat
-ST-Elevation Myocardial Infarction (STEMI)
-Atrial pacing of the pacemaker
 lOMoARcPSD|44249588

For simplicity, we did not consider the multi-labeled ECG classification problem in this study,
focusing only on the single-label, single-lead ECG classification problem. The labels (i.e.,
diagnoses) for these 14 classes were determined by multiple cardiologists in university hospitals
across South Korea. We attempted to classify all of the aforementioned diagnoses using the ECG
signal data. In this paper, through deep learning, we prioritized the classification of the most
fundamental rhythm type ECG (Normal, AF, APC and VPC), and for this purpose, we labeled the
rest of the most fundamental rhythm type ECG as “other”. In this study, our model was designed
to classify five rhythm classes (normal/AF/APC/VPC /other). The data proportions were as
follows: 43.30% (normal), 24.34% (AF), 11.71% (APC), 17.49% (VPC), and 3.16% (other).

We implemented a total of 158,892 (= 12 × 13,241, where 13,241 corresponds to the total number
of single-lead ECG datasets) 10-s-duration single-lead ECG datasets. Also, we used stratified
random sampling for split each single-lead ECG dataset into three different datasets, i.e., a
training dataset, validation dataset and test dataset, at a ratio of 64:16:20 and each class are split
to same ratio into each dataset; however, we fixed the random seed to enable direct comparison
of the results. The number of training samples for each single-lead ECG (Lead I, II, III, aVR,
aVL, aVF, V1, V2, V3, V4, V5 and V6) data was 8,473 (= 0.64 × 13,241). For validation and
testing of the ECG signals, 2,119 and 2,649 single-lead samples were respectively used to
validate and test each type of single-lead ECG data. Each single-lead ECG dataset comprised
ECG data that were collected over a period of 10 s at a frequency of 500 Hz. We changed the
data frequency to 200 Hz using linear interpolation (put the details to the appendix A). In
addition, min-max normalization was applied to normalize the range of ECG output values and
maximize DL model training efficacy.

Data Mining and ECG Retrospective Analysis

Several studies have illustrated how large-scale ECG data analysis can uncover useful insights—
such as the identification of inconsistent methods of measurement, overlooked abnormalities that
may suggest an underlying disease, or the impact of making a change in how ECGs are
processed across an institution.

Retrospective Analysis Often Reveals Errors in Initial Results

A recently published study in Cureus compared Bazett's QT interval formula (QTcB) with
Fridericia's formula (QTcFri) to determine the extent of difference in those identified as
"prolonged QT." Previous research indicates Bazett's formula is actually more sensitive to heart
rate, and most modern healthcare data management systems automatically use Fridericia's
formula in its place. After examining more than 44,000 ECGs retrospectively, researchers found
that 21% of all the ECGs corrected to normal QTc when Fridericia's formula was applied.4 This
equates to 57% fewer ECGs identified as having a QTc greater than 500ms, which is the
threshold for identifying patients at risk for drug-induced torsade de pointes (TdP).
 lOMoARcPSD|44249588

Obviously, a change of 57% impacting thousands of ECGs in this hospital-based study has
important clinical consequences impacting patient care, such as the administration or
withholding of medication. Consequently, when comparing ECGs over time, it is vital that a
consistent method be used, including the correction formula for correcting QT for heart rate.

Even without accounting for initial errors in ECG interpretation, a retrospective analysis may
yield previously overlooked information important for predicting risk for mortality. A study in
the International Journal of Cardiology reviewed ECG data from over 342,000 primary care
patients over a period of 10 years. Upon review of the ECG data, researchers found an
association between abnormal T waves (those with asymmetry, flatness, or notching) and
mortality risk independent of other patient-specific factors, such as heart rate, QTc, and other
baseline comorbidities.

Additional Benefits Help Workflow and Healthcare Costs

Retrospective analysis of ECGs may also be useful for determining cost-effectiveness and
optimizing workflow. A 2015 study in The Journal of Pediatrics examined clinical data from over
8,600 ECGs performed as part of athletic pre-participation evaluations (PPE) to assess children
and young adults' readiness to participate in sports. From 2005 to 2010, pediatricians selectively
performed ECG during the PPE based on suspicious cardiac symptoms, such as dizziness or
syncope. Only 0.5% of individuals receiving PPE alone were referred to cardiologists based on
clinical suspicion, but 13% of individuals receiving PPE and an ECG were referred to
cardiologists for further evaluation.

Evaluation of the same study's data years later further supports the use of ECG for diagnostic
purposes, as PPE alone identified cardiac disease with a sensitivity of only 44%. Within one year
of the PPE, cardiac disease was diagnosed in 0.5% of people receiving the test. However,
incorporating the ECG into a PPE helped identify underlying cardiac disease in 18% of
participants, allowing more patients to receive therapeutic procedures.

AI's Role in Evaluating Data

The expanding use of AI can also benefit clinicians seeking to analyze previously recorded
ECGs. The machine-learning techniques involved go beyond evaluating only a small portion of
the information contained in an ECG readout; rather, researchers are aiming to develop
algorithmic frameworks that support large-scale ECG analysis. Ultimately, this may help
physicians save time and effort when interpreting ECG findings and making treatment decisions.

Algorithms Increasingly Aid in Trend Identification and Diagnosis

Researchers created this type of algorithm and published their findings in Circulation:
Cardiovascular Quality and Outcomes with the ultimate goal of developing an automated,
scalable, and interpretable method of characterizing cardiac structure and diastolic function as
well as detecting and tracking disease via patient-specific ECG profiles.
 lOMoARcPSD|44249588

In the examination of standard 12-lead ECGs collected from 2010 to 2017, the model was
restricted to ECGs presenting normal sinus rhythm. Using continuous metrics, the AI approach
enabled estimation of the severity of structural abnormalities, which were verified against
reference electrocardiographic measurements from GE's MUSE Cardiology Information System.
The algorithm also helped to classify four example diseases (pulmonary arterial hypertension,
hypertrophic cardiomyopathy, cardiac amyloidosis, and mitral valve prolapse), even identifying
new ECG predictors for each disease.

Another machine learning platform, known as "rECHOmmend", was developed to analyze data
from any ECG system to predict various types of structural heart disease, including clinically
significant valvular disease, reduced left ventricular ejection fraction, and pathologically
increased septal thickness. Using over 1 million ECG traces combined with information about
each patient's age and biological sex, the platform demonstrated an overall positive predictive
value of 42% for clinically meaningful structural disease with 90% sensitivity and 73%
specificity. These findings, published in Circulation, could eventually be used to identify high
risk patients using only ECG data.