0% found this document useful (0 votes)
35 views16 pages

DH302 Notes

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 16

DH302

Created @October 31, 2023 9:46 PM

Tags

Diagnosis: The process of identifying a disease, condition, or injury from its signs
and symptoms. A health history, physical exam, and tests, such as blood tests,
imaging tests, and biopsies, may be used to help make a diagnosis.
Prognosis: The likely outcome or course of a disease; the chance of recovery or
recurrence.
Sign: something found during a physical exam or from a laboratory test that shows
that a person may have a condition or disease. Some examples of signs are fever,
swelling, skin rash, high blood pressure, and high blood glucose.
Symptom: A physical or mental problem that a person experiences that may indicate
a disease or condition.
Easing Processes -
1. Making medical care less expensive: approximately 30% of health care spending
may be considered waste
2. Reduces prescription errors
3. Expand access to care

4. Improve quality of care


eHealth - the use of information and communication technologies (ICT) for health to
treat
patients, pursue research, educate students, track diseases and monitor public
health.
Telemedicine - facilitation of healthcare delivery to a location that is physically apart
using
means such as the internet
Electronic Medical Records - An EMR is a legal electronic record of an individual’s
medical information and thus a focused narrower view of the patients’ medical history
Electronic Health records (EHR) - Health records can be collected as a part of a
continuum even when the individual is healthy. This comprehensive collation of
information is considered to be electronic health records (EHR)

DH302 1
Unlike an EMR, an EHR can span multiple healthcare organizations as also multiple
sub-systems including administrative, laboratory, radiology, pharmacy, pathology,
cardiology, pulmonology, dermatology, orthopedics, Dental, ENT, neurology,
nephrology, oncology
Radiology- The use of radiation (such as x-rays) or other imaging technologies
(such as
ultrasound and magnetic resonance imaging) to diagnose or treat disease.
Pathology - The branch of medicine that deals with the laboratory examination of
samples of
body tissue for diagnostic or forensic purposes.
Autopsy - An autopsy, also known as a post-mortem examination, is a specialized
surgical procedure used to determine the cause and manner of death.
Biopsy - A biopsy is a procedure to remove cells, tissue or fluid for
examination by a medical pathologist.

Cardiology - A branch of medicine that specializes in diagnosing and treating


diseases of the
heart, blood vessels, and circulatory system.

Electrocardiogram - An electrocardiogram records the


electrical signals in the heart
Pulmonology - A branch of medicine that specializes in diagnosing and treating
diseases of the
lungs and other parts of the respiratory system

Dermatology - the branch of medicine concerned with the diagnosis and treatment
of skin
disorders.

Orthopedics - The branch of medicine dealing with the correction of deformities of


bones or
muscles

Neurology - The branch of medicine that deals with the diagnosis and treatment of
disorders
of the nervous system
Nephrology - the branch of medicine that deals with the physiology and diseases of
the
kidneys.

DH302 2
Oncology - The study and treatment of CANCERS

Information management - The generation, collection, organization, validation,


analysis, storage, and integration of data as well as the dissemination,
communication, presentation, utilization, transmission, and safeguarding of the
information

An automated system that uses computer hardware and software to record,


manipulate, store,
recover, and disseminate data
Biomedical informatics - Biomedical informatics (BMI) is the interdisciplinary field
that studies and pursues the effective use of biomedical data, information, and
knowledge for scientific inquiry, problem solving, and decision making, motivated by
efforts to improve human health

1. Clinical informatics - First set of applications were focused towards finding


solutions
associated with clinical care and this is labeled as Clinical Informatics.

2. Public health informatics - Public Health Informatics is closely associated with


clinical
informatics, the only difference being, here the unit of focus is a population and
not an individual. Its scope includes disease surveillance, clinical trials for
vaccines, immunization registries as well as bioterrorism.

3. Imaging informatics - Imaging Informatics concerns itself with the acquisition of


images,
their conversion to digital form and their subsequent manipulation and analysis.

4. Bioinformatics - It includes the investigation of the representation of information


and the
processes involved in transmitting the information within biological systems,
beginning with the molecular level.

Genomics - study of genomes

Proteomics - Study of proteins


Metabolomics - study of metabolites

Transcriptomics - study of differential gene expression


Big Data - Big data is defined by the five V’s, Volume, Velocity, Variety, Veracity and
Value

DH302 3
Volume is referring to the total amount of data,
Velocity is the pace at which the data is generated,
Variety is different kinds of data generated within the same domain,
Veracity is a measure of how genuine the data is
Value is the quality of the data in terms of the expected outcomes of the data after
processing.

Immunology - Immunology is the branch of biomedical science that deals with the
response
of an organism to antigenic challenge and its recognition of what is self and
what is not.

Stroke Damage to the brain from interruption of its blood supply


Reperfusion therapies, are the only effective treatments to reverse the ischemic
changes

Drug discovery Informatics market will reach a market value of $ 6.3 billion by 2030.

Drug discovery informatics is defined as a method to evaluate and investigate large


amounts of data of biological origin. This data is derived from clinical research
Laboratories and utilizes state-of-the-art algorithms to analyze data associated with
the disease, the drugs used for its treatment, the genes and the pathways involved,
the protein pathways involved etc.

Pharmacokinetics - What the body does to the drug

Pharmacodynamics - What the drug does to the body


Quantitative structure–activity relationship (QSAR) and quantitative structure–
property relationship
(QSPR) models
OCR : the process that converts an image of text into a machine readable text format

Interoperability in healthcare refers to the secure access and integration of


electronic health data so it can be used to optimize health outcomes.
Fast Healthcare Interoperability Resources, also known as FHIR, is an open
source standards framework for healthcare data that builds off of a previous
standards framework called HL7. FHIR was created to make it easier for healthcare
data to move from one system to another.

Fragmented data - 20-25% of US healthcare spending or approximately $1 trillion


dollars is currently wasted, and 50%-75% of this waste could be eliminated by using
an updated and shared electronic medical platform.

DH302 4
Case Report: A research article that describes and interprets an individual case,
often written in the form of a detailed story.

Case Series: A case series is a type of medical research study that tracks subjects
with a known exposure, such as patients who have received a similar treatment, or
examines their medical records for exposure and outcome
Clinical trials: Clinical trials are research studies that test a medical, surgical, or
behavioral intervention in people

ABHA - Ayushman Bharat health account


Health Insurance Portability and Accountability Act (HIPAA) regulations

Protected health information (PHI) is the term used by HIPAA to describe any
information about an individual’s past or present health status or treatment that can
be used to identify them

1. The Privacy rule

2. The Security rule

3. The Breach notification rule

Ethical Challenges - current gap is the inability of policy ecosystem to adapt to the
rapid advancements that is happening in the digital healthcare space thereby
increasing the possibility of loopholes to be exploited.
Types of Health care data:

Clinical data - data points that are associated with the present and past health
status of the individual.

Disease registries: a special database that contains information about people


diagnosed with a specific type of disease

Psychological data: Mental, emotional, environmental and other behaviour factors


that are associated with the patient

Data associated with lifestyle: eating habits, exercise habits, smoking. several risk
factors with chronic conditions are increasingly being looked at.

Demographic data: age, gender, ethnicity, Place of birth and other such demographic
details
Sickle cell anemia
Genomic data: marked individual variability within a heterogeneous group of
population which

DH302 5
is driven by genetic influences as well as the interaction of genetic influences
with the environment

Epigenetics: the study of changes in organisms caused by modification of gene


expression
rather than alteration of the genetic code itself
Metadata: a set of data that describes and gives information about other data.

Descriptive metadata enables discovery, identification, and selection of resources.


It can
include elements such as title, author, and subjects.
•Administrative metadata facilities the management of resources. It can include
elements such as
technical, preservation, rights, and use.
•Structural metadata, generally used in machine processing, describes
relationships among various
parts of a resource, such as chapters in a book.

Patient identity metadata is useful to identify and select a unique patient from the
entire population of individuals on whom we have healthcare data.
Provenance metadata is the type of data set that indicates the origin as well as the
history of modifications that have been associated with this data.
Privacy metadata potentially provides information of the consent that the patient
has afforded for data sharing

Neuron
The basic unit of communication in the nervous system is the nerve cell (neuron).
The myelin sheath is fatty material that covers, insulates and protects nerves of the
brain and spinal cord.
Communication between neurons occurs at tiny gaps called synapses, where
specialized parts of the two cells (i.e., the presynaptic and postsynaptic neurons)
come within nanometers of one another to allow for chemical transmission
Multiple sclerosis (MS) is a potentially disabling disease, In MS, the immune
system attacks the protective sheath (myelin) that covers nerve fibers and causes
communication problems between your brain and the rest of your body. Eventually,
the disease can cause permanent damage or deterioration of the nerve fibers.
Epilepsy is a disorder of the brain characterized by repeated seizures. Absence
Seizure: A type of seizure that involves brief, sudden lapses in attention.
Health information technology or HIT is a system that is intended to acquire,
store, organize, analyze and finally potentially share the healthcare data that is

DH302 6
collected in any health care facility.

1. Resources – Resources are the frameworks that are necessary for the optimal
functioning of the health information system. Resources include but are not
limited to:
· the financial requirements to run the system
· logistic support to operate the system
· the personnel that operate the system
· the technology that is necessary to operate the system
· the regulatory and the legislative requirements that are specific to
this system.

2. Indicators: The health information system should be able to provide data of


various needs such as supporting doctor decisions, providing statistics for
managing services as well as assessing the impact of health policies

3. Data Sources:

Population level data sources can be defined as information coming from


censuses, civil and disease registries as well as population surface for defined goals.
Institution-based data sources are not only restricted to the hospital based
electronic records but also reports from occupational registries that record health
information of workers as well as work related injuries as well as records such as
food and agriculture production and distribution.
Surveillance data of both communicable and non-communicable diseases become
major data sources. Such surveillance data is both from institution based
surveillance (eg: emergence of antibiotic resistance in the hospitals by following on
culture antibiotic Sensitivity assays) as well as population based surveillance

4. Data Management: One of the fundamental principles is the principle of


‘minimum data set’ which decreases the total amount of information that needs
to be collected which by itself
improves data quality

5. Information products: Data by themselves do not add any value to our defined
goals and objectives unless the data is processed and analyzed generating
tangible information

6. Data use: Transforming data to information as previously mentioned contributes


to development of informed clinical decisions as well as broad policy level
decisions.

DH302 7
Cytokines are made by many cell populations, but the predominant producers are
helper T cells
Chemokines: Chemokines direct immune cells toward places in your body where
they can fight infection.
Interferons: Interferons signal cells to put up their defenses against viruses invading
your body. In this way, interferons “interfere” in the process that allows viruses to
replicate, or make more viruses once they’ve invaded a healthy cell.
Interleukins: Interleukins get their name from “inter” which means between and
“leukocyte,” which is another name for a white blood cell.
Tumor necrosis factor (TNF): TNF helps regulate inflammation in your body. TNF
also signals to immune cells that kill tumor cells.
Patient centered care
It is defined as the provision of healthcare that is in accordance with the patient's
preferences, values and needs
Types of Health informatics system:

1. Medical practice management (PM) software streamlines, coordinates, and


automates administrative tasks in doctor’s offices, hospitals, clinics, and other
healthcare organizations. The main users of the tool are non-medical employees
like receptionists and administrators

2. Electronic health record system:

It deals with electronic health information, which generally contains information on


patients’ medical history, allergies, laboratory charges, and more.

3. E-Prescribing Software
Software allow the treating physician to prescribe the medicines as well as the
medicine regimen electronically which is then delivered directly to the pharmacy
from which the patient can pick up their medicines.
e-prescribing software are decreased risk of misplacement of the prescription as
well as a comprehensive record of the medicines that have been prescribed to
the patient.

4. Remote Patient monitoring:

It can help monitor a patient’s glucose level or blood pressure from anywhere
worldwide. Thus, it results in improved patient care

5. Master Patient Index:

DH302 8
The Master patient index is a crucial part of the hospital management system as it
connects patient records from more than one database.
MPI creates an index of all medical records for a specific patient, which is easily
accessible by all departments.
Reduce duplication of patient records and avoid inaccuracy of information that can
result in unfair treatment

6. Patient Portal :

A platform to access health-related data using any device. It includes all the
information stored in an EHR, such as the patient’s medical history, treatments, and
previous medications.

7. Urgent Care applications:

type of health management information system that keeps track of patients that
might require immediate attention

8. Medical billing software: It is one of the most time-consuming processes and yet
one of the critical ones

Diabetes:

Diabetes is a serious condition where your blood glucose level is too high. It can
happen when your body doesn't produce enough insulin or the insulin it produces
isn't effective. Or, when your body can't produce any insulin at all.

The Receiver Operator Characteristic (ROC) curve is an evaluation metric for


binary classification problems. It is a probability curve that plots the TPR against FPR
at various threshold values and essentially separates the ‘signal’ from the ‘noise.’
Electronic Medical records:

The Receiver Operator Characteristic (ROC) curve is an evaluation metric for binary
classification problems. It is a probability curve that plots the TPR against FPR at
various threshold values and essentially separates the ‘signal’ from the ‘noise.’

Continuity of Care Documents (CCDs) are used to support handoffs from one
clinician to another. They capture what health standards organization HL7 calls a
“snapshot in time” - a standardized summary of the relevant clinical data for a
specific patient.
Patient centered care is defined as the provision of healthcare that is in accordance
with the patient's preferences, values and needs.

Pecker’s eight principle of Patient centered care ( Respect for patient’s


preference, coordination and integration of care, Information and education, Physical

DH302 9
comfort, Emotional Support, Involvement of family and friends, Continuity and
transition, Access to care)
Make shared decisions for treatments.

Metastasis - Development of secondary growth at a distance close to cancer


What is HbA1C? - average blood glucose level
Health Information System(HIS) - "generation of information to enable decision-
makers at all levels of the health system to identify problems and
needs, make evidence-based decisions on health policy and allocate scarce
resources optimally"

1. Medical Practice Management System - it streamlines, coordinates, and


automates administrative tasks in doctor’s offices, hospitals, clinics, and other
healthcare organizations.

The main users of the tool are non-medical employees like receptionists and
administrators. Appointment Scheduling, Insurance Claims.

2. Electronic Health Record Systems(EHR)- It deals with electronic health


information, which generally contains information on patients’ medical history,
allergies, laboratory charges, and more.

3. E-Prescribing Software - prescribe the medicines as well as the medicine


regimen electronically

4. Remote Patient Monitoring - It can help monitor a patient’s glucose level or


blood pressure from anywhere worldwide.

5. Master Patient Index - connects patient records from more than one database.
Reduce duplication of patient records and avoid inaccuracy of information.

6. Patient Portal - A platform to access health-related data using any device

7. Urgent Care Applications- keeps track of patients that might require immediate
attention.

8. Medical Billing Software- Besides patient billing, the software covers insurance
claims and verification, payment tracking, and processing

Diabetes Mellitus - It happens when your body does not produce insulin or insulin is
not effective.

Type 1 diabetes - Autoimmune reaction


Type 2 - body doesn’t use insulin and cant keep insulin at normal blood levels

DH302 10
Diabetic retinopathy is caused by damage to the blood vessels in the tissue at the
back of the eye (retina).
(Receiver Operator Characteristic (ROC) curve) - evaluation metric for binary
classification problems. higher the AUC, the better the performance.
Electronic Medical Records - contains the patient's records from doctors and
includes demographics, test results, medical history, history of present
illness (HPI), and medications:
Elements of Electronic Health records - Physicians, clinicians, Hospitals, Radiology
reports, Laboratory data, Insurers, Vital signs.

Key components of EHR - Demographic data, Financial details, health history,


follow up progress report, imaging reports, prescriptions
Basic health data standards - possibility of data sharing in predefined formats.

one has to avoid any ambiguity while transferring as well as interpreting the
information. interoperability
Benefits of healthcare data standards adoption - Data Integration, Improved
compliance, Improved compatibility, better clinical decisions, seamless financial
transactions.
Challenges - Fragmented data and Incompatibility

FHIR - Fast healthcare interoperability resources - Rules that govern the exchange
of electronic data between different information technology
system

C-CDA - Consolidated- Clinical Document Architecture- each patient encounter is


represented by a single document.
C-CDA is read only. It is a library of templates. as known as Continuity of care
document(CCD)
Continuity of Care Documents (CCDs) are used to support handoffs from one
clinician to another. They capture what health standards organization HL7 calls a
“snapshot in time” - a standardized summary of the relevant clinical data
for a specific patient
Problem with CCD - EHRs don’t actually store information in the CDA or C-CDA
format. (compatibility issues)
FHIR - it solves the problem with CCD. it allows users to process data with JSON,
permits flexible, atomic data sharing, its regulatory and Developer support are
unmatched.

DH302 11
Resources(atomic data) - fundamental building blocks of FHIR. instead of sharing
XML reports . FHIR permits requests for resources.

Renal system - Kidney stones (also called renal calculi, nephrolithiasis or


urolithiasis) are hard deposits made of minerals and salts that form inside your
kidney
Non-contrast computed tomography (NCCT) is a standard practice for
investigating flank pain
Organ Rejection - acute(first few months) ,Chronic rejection - (a year or time
beyond)

you will need to take medication (immunosuppressants) for the rest of your life to
keep your body from rejecting your new organ.
Clinical Decision support system - improve healthcare delivery by enhancing
medical decisions with targeted clinical knowledge, patient information, and other
health information.
Traditional CDSS - Patient’s current medications + Patient specific information +
Drug information
A typical CDSS contains three core elements: a base or data management layer,
inference engine or processing layer, and user interface

Knowledge Based CDSS - data is structured in the form of if-then rules.


Non Knowledge based CDSS - system learns from past experiences and finds
patterns in historical data ( ML models)

genetic algorithm - GAs adapt to a new task, producing a number of random


solutions and then iteratively evaluating and improving them until the most fitting
option is found.
Five phases of genetic algorithms - Initial population, fitness function, selection,
crossover, mutation

1. Fitness Function - The probability that an individual will be selected for


reproduction is based on its fitness score.

2. Selection - The idea of selection phase is to select the fittest individuals and let
them pass their genes to the next generation

after fitness we use selection


Roulette wheel selection: fitness/ total fitness

DH302 12
Event selection: It can make the algorithm more adaptive and robust, especially
when dealing with complex optimization problems. it also increases the complexity of
the algorithm.
Rank- grounded selection: it assigns selection probabilities to individuals based on
their rank or position in the population rather than their absolute fitness values

3. Crossover- a crossover point is chosen at random from within the genes

4. Offspring - created by exchanging the genes of parents among themselves until


crossover is reached

5. Mutation - Mutation occurs to maintain diversity within the population and


prevent premature convergence.

Termination - the algorithm terminates if the population has converged.


CDSS functions - Patient safety, alert fatigue. solution - prioritize critical alerts.
clinical management, negative impact on user skills . solution - set up systems,
without jeopardizing autonomy or being too ‘prescriptive’ and definitive

cost containment, financial challenges - solution: payers ay be more willing to


support CDSS.
QALY(quality adjusted life years)- a metric used to judge quality of life.
Gastroduodenal Ulcer - direct action of h-pylori bacteria on gastric epithelial cells
Health Promotion - enabling people to increase control over their
health. health promotion involves: personal agency, ecological overarching frame,
access and increasing engagement.
Global Systems for Mobile Communications Association (GSMA) 8.8 billion mobile
connections and 5.1 billion unique mobile device subscribers.
interesting about digital health technologies - reach hard to access populations, easy
and affordable, personalized to individuals
The gender gap merits specific attention. Women in lower- to middle-income
countries are 10% less likely than men (197 million women) to own a mobile and
23% less likely to access mobile services.
Engagement - improving walking, muscle strength and balance
Rewards - Healthy 365 (H365) is an app launched by the Health Promotion Board
(HPB) of Singapore to encourage users to adopt a
healthier lifestyle through games and rewards.

DH302 13
Benefits of Digital health promotion - Decreased upscaling costs, personalized public
health, Mammography recommendation.
WISDOM Study (Women Informed to Screen Depending On Measures of risk)
Type of comparisons in clinical trials -

1. Superiority trials: new treatment intervention (drug, psychotherapy) is


superior to (better than) the control condition.

2. Non- inferiority trials: treatment A is not worse than the treatment B. the new
intervention may be less costly, less invasive, and have less
side effects.

3. Similarity trials - By quantifying ‘Similarity’ using a tolerance range

Mammography - X ray of the breast


Nudging - : any aspect of the choice architecture that alters people’s behavior in a
predictable way

Choice personalization - concerns the options presented to the patient


Delivery personalization - refers to the method of nudging
Secondary use or privacy - misuse
Challenges of digital health promotion: Algorithm bias, cybersecurity risks, mis-
information. effects of influencing- limits of nudging and the right of free choice,
Minimizing social inequities through widespread accessibility and affordability
Digital twin in healthcare:
A virtual replica of a product or system is called a "digital twin” throughout its life
cycle. Digital twins provide learning, reasoning, and dynamic recalibrating for
improved decision-making using real-time data
It is a virtual representation of
a device that is interactive and continually collects data using software and
embedded sensors
we can accurately portray each person and perform simulations
Digital Twin does risk analysis on a variety of what-if situations and helps firms
maximize operational efficiency

DH302 14
Liver system -

Portal vein - Portal vein delivers blood from organs in your


abdomen to your liver for processing
Cirrhosis is a condition in which your liver is scarred and permanently damaged.
Hepatic Encephalopathy
Hepatic encephalopathy (HE) is a brain disorder caused by chronic liver failure,
particularly in alcoholics with cirrhosis, which results in cognitive, psychiatric, and
motor impairments.
Nonalcoholic fatty liver disease (NAFLD) is an umbrella term for a range of liver
conditions affecting people who drink little to no alcohol.
Ayushman Bharat digital mission : 27th September 2021.
Develop the backbone necessary to support the integrated digital health ecosystem
of the country. and Ensure a seamless exchange of health records enhancing the
quality of care, efficiency, accessibility, and affordability
HPR - A comprehensive repository of registered and verified healthcare
professionals across all systems of medicine
PHR - electronic application through which individuals/patients can maintain
and manage their health information

UHI unified health interface is an open network to enable interoperability among


digital health services.
Health Claims Exchange (HCX) is an open network to enable interoperability in the
health insurance ecosystem
Health Management Information System (HMIS) - patient health records, clinical
records, financial accounts, laboratory reports, medication, inpatient and outpatient
information

DH302 15
Why we use max pooling in CNN? to downsize the image and removing
invariances. it also adds translational invariance.
for predictive model of peptic ulcers Elastic net regularization favors the selection
of predicting parameters.
it selects six predicting parameters - age, baseline hemoglobin, presence of
gastric ulcer, gastrointestinal diseases, malignancies, infections
Colorectal cancer - cancer in colon

RLHF - reinforcement learning from human feedback


adapters - more efficiency by a smaller number of parameters required to
train, which is called adapters.
Current Procedural Terminology (CPT) codes are a standardized system of
medical codes developed by the American Medical Association (AMA) to describe
medical procedures, services, and treatments provided by healthcare
professionals.

DH302 16

You might also like