DICOM (Digital Imaging and Communications in Medicine) is an international standard used for storing, integrating and exchanging medical image data. DICOM is mainly used in radiology, cardiology imaging, x-ray, CT, MRI etc. and one of the widely used medical standards by clinicians in the world.

DICOM was established in 1982. Back in olden days exchanging medical images from one device/software to other was a cumbersome process because of different formats adopted by device. It involves conversion of source image into target device specific format. Also storing images with patient information. DICOM created to solve these compatibility issues. The main purpose of this standard is to enable interoperability among various medical devices/software’s to create, exchange and view medical image files without any specific conversion.

This standard is maintained by a committee comprising of 25-30 companies, 10-12 user organisations, and 6-8 general interest members. NEMA (National Electrical Manufacturers Association) holds the copyright and oversees day to day operations, publications and legal problems. The standards committee is responsible for developing and voting proposed standards and adopting policies and procedures.

The ACR(American College of Radiology) oversees the technical and medical instructions. DICOM is an integral part of IHE(Integrating the Healthcare Enterprise). DICOM standards committee jointly works with HL7 group and ISO’s TC 215.

Several formats available in market to represent images namely JPEG, PNG, GIF etc. This formats doesn’t allow to store extra information and provide the ability to search inside the image for particular pattern. Each medical image should be associated with patient information and other required extra information specific to the device from which it is captured. DICOM allows to store patient and extra information and search functions. This way when exchanged other clinicians can identify patient from the image data.

DICOM standard consists of specifications for

  1. Defining images including its structure : Basic format how the image data stored with extra information
  2. Network protocol: Specifications for network services
    • Archive service – to search an image in the system
    • Print service – to allow access to the network printers
    • Modality work list service – To allow to download latest work lists with patient demographic data from other systems
  3. Formats for exchanging – how the two parties agree on the format for exchanging with compression and other information.
  4. Conformance specifications: Each independent device or software that supports DICOM must conform to the minimum set of conformance specifications.

A DICOM file contains

  1. Header
  2. Data elements

Header is repeated only once and is sometimes optional. It consists of 128 preamble and 4 byte prefix. Data elements are repeated and each one associated with a tag. Group of data elements forms a data set and are ordered by data element tag number.

Each of this data element consists of

  1. Patient information with name, sex, identification number
  2. Device parameters, scan type
  3. Image information such as resolution, windowing, width and height

Recent National E-Health implementation added additional identifiers IHI, HPI-I and HPI-O can be used for an individual and provider on top of existing identifiers such as Medicare, Insurance etc. Different identifiers are used for different purposes for example insurance identifier is used when sending  insurance information to identify specific individual, IHI number used when accessing PCEHR, organisation may use one Medicare number for billing purposes and other for administrative purposes.

Australian standards provided new set of guidelines in 2014 to accommodate these identifiers in the health systems.

Every identifier should be associated with the following when stored 

  • Identifier designation : This refers to the actual identifier code

Example: IHI number or Medicare number

  • Identifier Issuer:  Name or HPI-O of the issuing authority

Example: Medicare, Centrelink, 8003621566684455

  • Identifier usage : The purpose of using the identifier

Complete list of identifier usages

110 – Individual Healthcare Identifier (IHI)
112 – Healthcare Provider Identifier—Individual (HPI-I)
113 – Healthcare Provider Identifier—Organisation (HPI-O)
114 – Virtual smart card identifier (CSP)
120 – Family Identifier
200 – Billing Identifier
300 – Business or Individual Taxation or Social Security Identifier
400 – Special Service Identifier (e.g. diagnostic services)
410 – Laboratory Services
420 – Radiology Services
480 – Other Diagnostic Services
500 – Individual Provider Identifier—other than the national number
600 – Organisational Provider Identifier—other than the national identifier
700 – Professional Registration Identifier
800 – Other
900 – Unreliable

  • Identifier Usage Start Date: Start date of the identifier usage, the date on which using this identifier began or will begin.
  • Identifier Usage End Date: End date of the identifier usage, the date on which using this identifier ended or will end.
  • Identifier Status :  Status of the identifier

Examples: Active, Inactive

  • Identifier Group : Group of the identifier

Examples: F (Family), T(Treatment), O(Other)

I think it is essential for all the health systems in the current market used by providers (practice management, EMR, EHR etc) to accommodate these identifiers to meet the identification standards. It will be interesting to see how the existing health systems such as ZedMed, Communicare etc. conforms these standards.

Healthcare identifiers (HI) Service developed by the federal, state and territory governments that uniquely identifies healthcare providers and consumers. These identifiers are important building blocks of E-Health in Australia and used to enable PCEHR system and are used on medical documents, patient wrist bands, tokens etc.

The HI service is mainly used by

  • public and private sector hospitals
  • general practice
  • clinical specialist
  • community health
  • healthcare administrators
  • allied health
  • aged care settings

Healthcare identifiers are categorised into the following

  1. Individual Healthcare Identifiers (IHI)  – For individuals receiving health care
  2. Healthcare Provider Identifier – Individual (HPI-I) – For healthcare providers and personnel involved in patient care
  3. Healthcare Provider Identifier – Organisation (HPI-O) – For organisations that deliver healthcare (hospitals or medical practices)

Further IHI is classified into the following

  1. Verified – Evidence of Identity(EOI) completed through TDS(Trusted Data Sources) or HI service
  2. Unverified – Not verified by EOI process
  3. Provisional – Not known to healthcare facility and expires in 90 days
  4. Deceased – Indicates death of an individual
  5. Retired – IHI has been inactive for 90 days and Fact of Death Data received from Registrar of Births, Deaths and Marriages.

Each identifier is a unique number adhered to ISO7812: AS 3523.1&2-2008 standards and is of 16 digits in length. It doesn’t contain any identification information such as age, location etc. and never be re-used.

The first 1-6 digits contains issuer identification number (for example medicare) and 7-15 digits contains a unique reference number that identifies an individual. The last digit 16 is reserved as check digit which can calculated from issuer identification and individual reference number.

The main functions of a HI service are

  • Allocating IHIs, HPI-Is, and HPO-Is,
  • To allow authorised users  to search, retrieve and validate IHIs, HPI-Is, and HPO-Is
  • To allow authorised users to maintain and publish certain data (allowed only with HPI-Is and HPO-Is)
  • To provide digital certificates for access
  • Retiring identifiers

E-Health software systems engaged in health care uses HI service to issue, assign and maintain national healthcare identifiers for consumers and providers.  These systems access HI service directly or indirectly ( depends on the other systems which have direct access to HI). All the systems should undergo conformance testing(CCA) performed by NATA recognised laboratories to prove it supports clinical safety, security and privacy.

Before discussing about algorithms, I will discuss some frequent terms used in DNA analysis. I’m not a biological student but one who uses algorithms  for DNA mappings, sequencing related analysis need to understand frequently used terms around DNA

  • Amino acids
  • Proteins
  • Nucleic acids
  • Genome
  • Transcription
  • Translation
  • DNA replication

Amino acids

Amino acids are chemical compounds which are building blocks of proteins and act as intermediates in metabolism. Protein characteristics always dependent on the amino acids precise content and their sequence. It means each protein has different amino acid contents and sequence. Also the chemical properties determine protein biological activity. To understand a protein structure and its stability (protein folding), it is essential to understand amino acid structure and chemistry first.

There are 20 types of amino acids within a protein 10 of which can be produced by human and other 10 are supplied through food. Human body doesn’t store excess amino acids (those available from food) like fat and starch so they should be available through food daily.

Recent study findings on insulin resistance for diabetes type 2 patients revealed that lipids and Branched-Chain amino acids (BCAA) work together to promote metabolic diseases. The presence of BCAA related signature is predictive of incidence, progression and remission of diabetes and insulin resistance.


Proteins are complex molecules that are made up of smaller unit’s amino acids as a long chain. Proteins can be categorized into types based on their function, few of the types are

  1. Antibodies – attaches to virus to protect the body
  2. Enzymes – Assist in the formation of new molecules from the DNA information
  3. Messenger – to transmit signals between cells, tissues and organs
  4. Structural component – Used for cell structure and support.
  5. Transport/storage – Carry atoms and molecules around the body
  6. Proteins can synthesised using two ways
  7. Biosynthesis – These are synthesised in cytoplasm from encoded gene instructions
  8. Chemical synthesis – These are synthesised chemically with peptide synthesis

A simple good example is identifying antibodies in the blood for determining HIV existence in the patient’s body

Nucleic acids

Nucleic acids are essential molecules of life. They include DNA (deoxyribonucleic acid) and RNA (ribonucleic acid) which are made from nucleotides. Each nucleotide consists of 5-carbon sugar, a nitrogenous base and one or more phosphate groups and categorised as DNA or RNA depending upon the sugar component. If the sugar component is deoxyribose, then it’s a DNA or if sugar component is ribose, then it’s a RNA.

DNA consists of two long strands of nucleotides which are anti-parallel in nature (opposite direction)

RNA plays an important role in protein synthesis and is divided into three types

  1. Transfer RNA (tRNA)
  2. Messenger RNA (mRNA)
  3. Ribosomal RNA (rRNA)

DNA and RNA contains sequence of genetic instructions that are essential for encoding cells, organs etc.


A Genome is encoded DNA genetic material of an organism which contains all the information needed for building and maintaining the organism. It is estimated humans contains more than 3 billion DNA base pairs in all the cells that have a nucleus. A genome contains set of instructions needed to build cells. Each one will have two types of

Genome sequencing and matching helps the analysts in predicting and matching like DNA. It is nothing but decoding genetic sequence in the form four letters A C G T. With newer technologies, sequencing became very cheaper when compared a decade ago.

Scientists uses genome compositions to study evolution history of genomes by comparing the proportion sizes of repetitive DNA and non-repetitive DNA.

Transcription is a process of creating a copy of mRNA (messenger RNA) from a DNA gene sequence. This mRNA enters cytoplasm after leaving cell nucleus. Cytoplasm direct protein synthesis according to encoded instructions stored in mRNA.

DNA contains two strands sense and antisense. mRNA is actually a single stranded unlike DNA and a compliment of antisense strand of DNA (template strand).

Transcription is done with the following steps

  1. Pre-initiation –
  2. Initiation
  3. Promoter clearance
  4. Elongation
  5. Termination

There is a process called reverse transcription in which RNA is transcribed into DNA (opposite of DNA to RNA). This behaviour usually found in viruses such as HIV etc.


Translation is a process of decoding and translating instructions from messenger RNA (mRNA) to direct protein synthesis. It converts gene sequence to amino acids sequence with the help of ribosomes and transfer RNA which in turn forms proteins.

Translation is done three steps

  1. Initiation
  2. Elongation
  3. Termination

Detailed steps involved in translation

  1. DNA transcribes genetic information by creating mRNA
  2. mRNA leaves cell nucleus and enters into cytoplasm
  3. mRNA carries genetic instructions from chromosomes to ribosomes
  4. Ribosomes assembles and translates genetic information to sequence of amino acids provided by tRNA
  5. A protein is formed based on amino acid sequences

DNA replication

DNA replication is process of replicating two identical DNA helices from the original DNA helix. This process is required in every living organism to carry out the following functions

  1. To build and regulate the cell from the encoded genetic information
  2. Genetic information transmission from one generation to other

The replicated helices are called as daughters and original DNA helix is called as parent. Parent strand is divided into two strands and a complimentary strand is created for each separated strand thus forming a daughter. This process is called as semi-conservative because each daughter contains one parent strand and one complimentary strand of parent strand.

By studying DNA replications, researchers were able to find the relations between certain disease behaviours. A recent example of this type of study is presence of DNA replication stress in human cancers. Human cancer is characterized by genomic instability. From the study, it was proved oncogene-induced DNA replication stress rises genomic instability in human cancers (increases deletions in common fragile sites). As a result genome copy number changes.

In the next part, I’ll discuss various algorithms and tools used for DNA analysis.

An EMR is an electronic health record system that keeps track of patient’s medical history specific to particular provider. It’s a digital version of paper charts maintained by providers. People who are in patient care(doctor, clinician, radiologist, pharmacist etc.) will access EMR to view and update medical records frequently.

An EHR is an electronic health record system that keeps track of patient’s medical history from more than one providers.

A Personal health record (PHR) is a tool used to collect, track and share past and current about your health or the health of someone in your care.

EHR and PHR are similar in nature (conforms to interoperable standards) but with some differences. EHR fully managed by providers whereas PHR is fully managed, shared and controlled by patient and can aggregate data from multiple sources. EHR’s are available to multiple providers without patient consent whereas PHR is available only on patient’s consent.

PHR are of two types based on who owns and controls the data, tethered PHR and interoperable. Tethered PHR is linked with EHR of multiple providers whereas interoperable PHR is independent of health providers and are more specific to the patient.

Hospitals maintains patient’s health records with EMR software which are fully controlled and managed by them and are specific in nature. This type of record management unless implemented part of EMR software have the following limitations when addressing patient’s needs. Most EMR’s available in the market doesn’t provide these features and are expensive to develop.

1) Lack of co-ordinated care of patient

The people engaged in patient care often have to access latest updated patient’s information for quick diagnosis and preventive medication. This information may come from various sources such as health devices, external health providers, and care team (doctors, clinicians, and radiologist).

2) Lack of sharing patient’s health records with other health providers

When patient attends another health provider, sometimes patient’s medical history from various hospitals attended is required to diagnose medical conditions.

3) Lack of patient’s ability to access medical history at any time

Patient have to access last hospital visit information and view medical history from time to time.

4) Lack of access to patient’s medical histories from different hospitals at one place

Patients may visit more than one hospital in their life time. With each visit, patient medical information is recorded. These records are not easily accessible from each provider to aggregate and store them at one place.

5) Lack of real time health monitoring updates from patient and care team

Often patient have to be monitored for chronic conditions with health devices such as blood glucose monitor, blood pressure monitor, heart rate monitor, pedometer etc. for diagnostic purposes.

Why PHR is preferred than EHR for addressing patient information needs inside and outside hospitals

Hospitals when addressing patient information needs to overcome the EMR limitations, usually can adopt EHR or PHR. EHR is specific to one or more providers and doesn’t provide interoperability with other hospitals. Still a patient can view and schedule appointments with EHR but restricted from doing any updates to their health information records.

I studied some research papers of some EHR implementations in hospitals, the results are as follows

· Hospitals were more likely to have implemented EHR if they were Larger Institutions, major teaching hospitals, Part of a larger hospital system, located in urban areas, or had dedicated coronary care units.

· No meaningful relationship was found between adoption levels of EHR and Ownership Status of the hospital in terms Public vs Private hospitals.

· Most commonly reported barriers were: inadequate capital for purchases (74%), concerns about maintenance costs (44%), resistance from physicians (36%), unclear return on investment (32%), and lack of staff with IT skills (30%).

· Among hospital that implemented EHRs, these factors were mentioned as having positive effects: additional reimbursement for EHR use (82%), financial incentives for adoption (75%), availability of IT implementation technical support (47%), objective third-party evaluation of EHR products (35%).

· Although adoption levels were low, a significant proportion of hospitals reported to have already used electronic format of laboratory and radiologic reports, radiologic images, medication lists, and some decision-support functions.

· Interoperability issues can also be noticed as barrier. Lack of health information exchange between hospitals has reduced the potential value of these systems.

· From a policy perspective, creating incentives for use of EHRs and creating disincentives for not using these systems can be a potential facilitator.

· Finally, it is important to notice that this study was focused on adoption levels of EHRs in US hospitals, and did not study the actual effectiveness of these systems.

PHR addresses patient needs more than EHR. It helps the patient to keep latest up-to-date information collected from various sources at one place.

Based on my study on PHR case studies, some of the PHR findings are

· Laboratory test results were the most commonly used feature. Scheduled visits, medical conditions, and current medications were among other frequently accessed features.

· Test results were populated to PHR with a delay of ten days to allow clinician review prior to patient access. However, it is concluded that patients want results to be simultaneously populated in their PHR.

· In terms of clinical information, Provider Profiles were the most accessed pages.

· More than 80% agreed that the PHR helped them manage their medical problems, get better prepared for their scheduled visits, and take charge of their healthcare.

· Although more than two-thirds of patients agreed the information was accurate and complete, a significant number of patients believed inaccuracies in current medications (25%) and medical conditions (32%).

· It is observed that patients desired PHR as a convenient tool of information exchange, and found it useful, particularly for laboratory test results, medication list, medical conditions lists, and links to information resources.

Why Health Vault PHR

There are several PHR solutions available in the market. Currently Health Vault PHR is the best PHR solution among the other PHR solutions. My selection of Health Vault solution is based on the following reasons for the evaluation

· Free with no licence fee but with affordable integration costs

· Bi directional (two way) updates with EMR

· interoperable with standards

· Cloud based scalable web solution

· Accessed through various mediums and devices such as tablet, mobile, health device, web services.

· Integrated with 20+ health devices

· Secure communication between parties

· Patient centred with controlled access

· Extensible with custom applications and services

· Customizable local storage

· Pluggable authentication providers such as Open ID providers specific to hospital

I decided to use triangular model evaluation framework for the evaluation of Health Vault application for use in hospitals.






Refers to hospital


Provider user

Refers to individual service providers in the hospital such as doctor, clinician and radiologist



Refers to Health Vault PHR


Patient needs

Refers to patient needs inside and outside hospital

I used some ratings based on functionality and accessing information in each criteria

1) Fully implemented 2) Partially implemented 3) Not implemented



Provide functionality to import or export Continuity of care document (CCD) and Continuity of care (CCR) documents


Secure authentication


Accessible via various mediums and devices (tablet, mobile, web)


Adheres to privacy principles


Ability to view and store medication history


Ability to view and store DICOM images


Ability to view and store family details


Ability to view and store miscellaneous files


Ability to view and store Allergies data


Ability to view and store immunization records


Ability to add and maintain health devices used for monitoring


Ability to maintain health profiles


Ability to share and print profiles


Ability to view and store procedures


Ability to view and store various health measurements


Ability to view access and update history


Ability to view and store family details and contacts


Ability to share and import information through email


Health Vault case studies

I have studied case studies of some hospitals using EMR software and integrated with PHR applications. The underlying principles and architecture is same for every PHR. Thus, I collected and studied some research papers on PHR general use in hospitals not specific to Health Vault.

Few hospitals who integrated Health Vault PHR with their EMR software as listed below

1) Maudsley NHS Foundation trust , UK

Maudsley foundation trust is the largest mental health service with over 35000 patients in Europe. Maudsley implemented a project called MyHealthBox which includes Health Vault integration to provide Summary Care Record to all patients

2) Mason General Hospital, WA, US

MyMasonHealth is a patient online solution implemented at Mason general hospital which integrates Health Vault with existing EMR using Microsoft Health Vault community connect to engage patients with improved experience in emergency rooms and clinics.

3) Brooks rehabilitation, Jacksonville, Florida, US

Brooks rehabilitation implemented a project Brooks Health Connect which integrated Health Vault with existing EMR. At Brooks, 90% of patients comes from referring acute care hospitals and are placed in various care settings within first 60 days. To improve patient experience and reduce the times between the team, Continuity of care solution is required to facilitate.

4) Steward Health Care System , Boston, US

5) Lucile Packard Children’s Hospital , Stanford, US

Lucile Packard Children’s hospital is a major referral center for severely ill children. The patients who visit the hospital are referred by Primary care provider outside LPCH are at 90% of overall visits. The remaining 10% patients visit directly to the hospital and its clinics. Hospital receives patient referrals from over 30 states. LPCH implemented an interoperable PHR like Health Vault to overcome the issues related to sharing and co-ordinated care.

6) PeaceHealth – St. Joseph Medical Center, Bellingham, Washington

Comprehensive preregistration and discharge process using HealthVault community connect.


Some concerns below related to usage of Health Vault PHR (or any general PHR) in hospitals, most of them addressed by industry expects

PHR records may not be accurate with current health status

I agree patient can modify health records which may not be true with current or past health history. Hospitals should not fully depend on PHR information. The main use of PHR is to provide up to date information of patient’s last visit and scheduling appointments. In my study, all the hospitals who integrated Health Vault PHR already implemented bi-directional updates with EMR and health vault. This means anything that is recorded in EMR will be synced to Health Vault and vice versa. On patient’s next visit, doctor access patient information from hospital EMR not Health Vault directly.

But if the patient visits another hospital (not linked to previous hospital) by modifying health vault information which is rare and if there is no record that is pointing to previous hospital, new hospital may fail in collecting previous medical history and may depends on Health Vault information.

Patient’s privacy

Health Vault or any other interoperable PHR is a patient controlled health record (PCHR). Patients can allow or deny access to their health records to any provider user at any time. Every hospital access is different and acted as a separate client for accessing the data.

Information Security

How safely and securely information communicated between various stake holders through Health Vault.

The communication between Health Vault and other parties is always secured and encrypted using industry standard protocols. Health Vault communication mainly done by patient, EMR system and provider users through https protocol and email.

Health Vault is a cloud deployed solution accessed by anyone using secure hypertext transfer protocol (https) through a web browser. Https is a standard secured protocol used by everyone and is done through digital certificates. Encryption and decryption done through private and public keys. At one end, the information is encrypted by a private key and other end the information is decrypted by public key. Client apps and services also have to do communication using https protocol to connect to Health Vault cloud.

Email is the other form of communication used by provider users. This communication is done using direct messaging protocol. Direct messaging protocol is a standards protocol which encrypts and secure a message. Each health vault user is allocated with a direct project email address when registering a health vault account. Currently Microsoft is offering this feature only for users registered in US instance.


Rebecca Todd, Maudsley launches Health Vault based PHR, 2012, E-Health insider (UK),

Arash Anoshiravani, Gregory Gaskin, Ed Kopetsky, Christy Sandborg, and Christopher A. Longhurst, Implementing an Interoperable Personal Health Record in Pediatrics: LessonsLearned at an Academic Children’s Hospital, 2011, J Particip Med

Health Systems Choose Microsoft HealthVault Community Connect to Improve Care Coordination, Engage Patients and Families, 2011, Microsoft

Lingxia Liao, Min Chen, Son Vuong, Xiaorong Lai, Novel Web-enabled Healthcare Solution on HealthVault System, 2012, Dept. of Computer Science, Univ. of British Columbia, V6T 1Z4, Canada

Bruno Alvesa, Henning Müllerab, David Godela, Omar Abu Khaleda, Michael Schumachera, Interoperability prototype between hospitals and general practitioners in Switzerland, 2010, University of Applied Sciences Western Switzerland (HES-SO), Sierre & Fribourg, Switzerland and Medical Informatics Service, University & Hospitals of Geneva, Switzerland

Nhan V Do, Rick Barnhill, Kimberly A Heermann-Do, Keith L Salzman,Ronald W Gimbel4, The military health system’s personal health record pilot with Microsoft HealthVault and Google Health, 2011,J Am Med Inform Assoc

Care Coordination Approaches with Microsoft Technologies, 2011, Microsoft

New Microsoft Health Solution Connects Hospitals with Patients and Referring Physicians to Improve Coordination of Care, 2010, Microsoft

Microsoft Health Vault https://www.healthvault.com/au




A Personal health record (PHR) is a tool used to collect, track and share past and current about your health or the health of someone in your care.

PHR’s are categorized into two types based on its data ownership and control

1) Tethered PHR

  • Linked to EHR of multiple providers
  • Provider owns and manages patient information
  • Extension of the provider’s health system EMR
  • Allows the patient to view information stored in EMR
  • Allows the patient to communicate via email or online appointment systems
  • Patients cannot share information electronically with outside providers
  • Patients cannot import and integrate their data from external sources

2) Interoperable PHR

  • Patient owns and manages his own health information
  • Stored in a central repository, accessible to all on consent
  • Electronic linkages with multiple providers, health systems and pharmacies
  • Integrated with provider’s health system EMR
  • Easily accessible and controllable
  • Patients can share information electronically with outside providers
  • Patients can import and integrate their data from external sources