- This topic has 22 replies, 13 voices, and was last updated 1 month ago by
.
-
Posts
-
-
-
In public health settings, I think COVID-19 data is one of ‘big data’, as the characteristics of data fits into 5Vs:
1.Volume: data provides large numbers of patients’ records such as demography, lab testings, treatments, vaccinations, social-economic, patients-related families’ health history records, etc. The more patients recorded, the bigger size of the data.
2. Velocity: real-time data or daily data reported to public health authorities and different organizations including WHO regarding COVID-19 updates.
3. Variety: Data provides different kinds of data including structured (eg. lab results), unstructured (eg. contact tracing, patient’s records, etc.)
4. Value: the data provides meaningful insights to health authorities to better respond outbreak in targeted areas, as well as strategy to prevent outbreak, and early diagnosis for the outbreak.
5. Veracity: the data can provide the health care providers/health authorities to make better decisions on disease outbreak response.Outside health-care settings, I think social media data such as Youtube, Instagram, Twitter, Facebook as well as Banking systems can be considered as big-data as it fits with 5Vs (at minimum), characteristics of ‘big-data’.
-
I think treatment data of HIV patients who take Anti-Retroviral Therapy (ART) can also be considered as big data as it also fits the characteristic of 7Vs as follows:
1. Volume: Over one hundred thousand of patients in national ART programs with decades of clinical records, viral load tests, CD4 counts, ART drugs histories and clinic visits due to lifelong nature of treatment.
2. Velocity: Rapid reporting of HIV related commodities consumption across nationwide can help supply chain management.
3. Variety: There are structured data such as drug prescriptions and lab results and also have unstructured data like counseling transcripts and clinical notes.
4. Veracity: Errors in manual data entry, self-reported adherence bias and incomplete demographic details.
5. Value: Helps to predict treatment failure and prevent drug resistance and improve retention in care, reduces mortality, and improve supply chain management.
6. Variability: Patients have different viral load suppression (VL) patterns (some achieve suppression in 6 months, others take longer) and VL machines have different levels of undetectable viral load count depending on manufacturer. Moreover, intervals of follow-up visits vary depending on the clinical status of the patient.
7. Visualization: Dashboards showing demographic and risk factors of new patients to take preventive measures, early warning indicator for pharmacovigilance monitoring, attrition rate for quality of care and drug consumption for supply chain management.
-
-
I think that EPI data is big data and it fit the 7 Vs as follow:
1. Volume is very large in EPI data because it collects the millions of under five children in country
2. EPI data have structured and unstructured data such as frequency of dosage, regular or catch up vaccination and AEFI incidence.
3. Velocity is also important one in EPI data, because if there is outbreak, the urgent rapid assessment of immunization status is necessary for PEP response.
4. For veracity, we have to check immunization history if the children are presented with vaccine preventable disease. EPI data is collected by the health staff by using the hospital records so it is accurate and reliable.
5. We can present the decision makers with visualization from EPI data. It will help to monitor vaccine coverage and diseases elimination programs.
6. Value: EPI data help to trace the children who are not receiving the full vaccine dosage.
7. It is vulnerable data because all confidential information of children are recorded.-
Hi Ama,
Thanks for sharing your opinion about EPI data. Your mention of Variety and Veracity shows the complexity of EPI data, since it includes both structured (dosage, coverage rates) and unstructured information (adverse events following immunization), and is collected by trained staff using verified hospital records. This helps ensure reliability for disease monitoring. It is good to know this process. -
Hi Hteik! Thanks for bringing up the EPI data. I agree with your explanation of the 7V characteristics of EPI data. I also would like to add that EPI data can provide more values, such as being used for epidemiology research, economic studies in the cost-effectiveness of vaccination programs, and social sciences e.g. to understand the barriers of vaccine uptake. Therefore, it is valuable not only for clinical use but also for guiding broader health policy.
-
-
In my opinion, Healthcare data could be considered to be fit in with seven Vs.
1) Volume: it is huge amount of data in a hospital. there might be million of pts records per year
2) Velocity: It is important to get real time data as well to use so timely updated is essential to upload.
3) Variety: There is also different type of data as pt record, lab result, stock data, financial data, logistic data with structured or unstructured data. That various type of healthcare data are needed to use in generating the impact of health care services.
4) Value: Data is precious when they use for multiple perspectives example value for money analysis, disease trending, medical research, making policy
5) Validity: It is one of core characteristic of data to record accurate data in health care. It means correct dosage of medicine, lab result, diagnosis etc.
6) Veracity: Health care data must be trustworthiness, good quality to apply in different purposes.
7) Visualization: It is also required to capture and generate the load of cases, financial growth, investment amount. Those visualized figures will help the management level to oversee the prominent points of health care setting to provide decision making for improvement. -
-
My opinion for an example of Big Data in healthcare is the Electronic Health Records (EHRs) combined with real-time patient monitoring data. For instance, hospitals now collect huge volumes of information such as laboratory test results, imaging files, medical reports, prescriptions, wearable device data (like heart rate, oxygen level, or physical activity), and patient histories. When combined, this creates a massive dataset that can be analyzed to improve diagnosis, predict disease outbreaks, and personalize treatments. This can be helpful in reducing patients’ unnecessary visit to the clinics.
Characteristics of this data according to 5Vs could be:
1)Volume: Healthcare systems generate large amount of data daily from EMRs, medical imaging, lab tests, wearable devices, and from billing systems. There is certainly large variety of data coming from different sources, in different formats.
2)Velocity: Data is generated rapidly in real-time—for example, continuous heart rate and oxygen saturation monitoring from ICU patients, and new wearable sensors helping track patient’s health trends that can be monitored by the doctors.
3)Variety: Data comes in multiple forms such as structured (lab results, demographics), semi-structured (sensor logs, prescriptions), and unstructured (doctor’s notes, CT scan images).
4)Veracity: Data quality can vary; some records may have errors, duplicates, or incomplete entries that require validation.
5)Value: When analyzed, the data can provide insights for better patient care, early diagnosis, and public health decisions. -
-
In my opinion, telecom data is one of the big data because it includes a massive dataset from millions of users, such as call records, billing history, location data, mobile data usage, etc. The characteristics are as follows:
Volume: A massive amount of data is generated daily by millions of users.
Velocity: The data is produced and processed in real time, such as call logs and mobile data usage.
Variety: The dataset includes multiple data types, such as structured (billing records) and unstructured (customer feedback).
Value: Information is important for improvement and detecting fraud.
Veracity: Accuracy and reliability of data are important for billing, location data, and network management. -
An example of big data is data from smartwatches. The characteristics of this data that fit into 5Vs are the following:
Velocity – Smartwatches can generate and transmit data in real time.
Volume – There are millions of smartwatch users in the world, generating a very large amount of data every day.
Value – Smartwatch can generate valuable health data which can be used in monitoring and predicting health outcomes of a patient.
Variety – Different types of data can be obtained from smartwatches. By using sensors, smartwatches can collect data such as heart rate, oxygen saturation, fitness, level, and sleep quality. The location of the wearer can also be determined using GPS.
Veracity – According to studies, accuracy of ECG measurement by smartwatches ranges from 65% to 99% while the accuracy of saturation of peripheral oxygen (SpO2) measurements ranges between 90% and 96%. Cardiac arrhythmias can also be predicted with an average accuracy of 97%.Reference:
Köhler, C., Bartschke, A., Fürstenau, D., Schaaf, T., & Salgado-Baez, E. (2024). The Value of Smartwatches in the Health Care Sector for Monitoring, Nudging, and Predicting: Viewpoint on 25 Years of Research. Journal of medical Internet research, 26, e58936. https://doi.org/10.2196/58936 -
-
I’ll give the example of big data in chronic kidney disease. Big data in the nephrology field can involve not only national databases such as the renal registries in the U.S., UK, Europe (ERA), etc, but also data collected through EHR/administrative claims, clinical trial registries, mobile health devices, and molecular databases. These big datasets fit the 5V framework as follows:
1. Volume: Data are collected from multiple large-scale sources as mentioned above.
2. Variety: Includes structured data such as laboratory values and diagnosis codes in registries, and unstructured data such as radiographs and pathology images.
3. Velocity: Refers to the speed of data generation and collection, which is accelerated by digital technologies. Wearable devices also contribute to real-time data generation.
4. Veracity: Data may be inconsistent and not fully standardized, but its trustworthiness still depends on the accuracy of diagnoses and laboratory values generated by professionals.
5.Value: This data has the potential to significantly improve the management and outcomes of chronic kidney disease.
-
-
I think EHR data from hospitals is big data. Because it has detailed information about patient’s profile, lab results, imaging results, medication history, follow-up information, admission information, allergy status, medical billings, surgical procedures and detailed progress notes. All the data is interlinked, and it can be retrieved anytime to make real-time decisions.
Volume: As I mentioned above, it has all detailed information for each patient across multiple years.
Velocity: Patient’s data can be accessed anytime when needed and can assist in real-time decision making and improve patient outcomes.
Veracity: Ensure data accuracy and consistency (by data validation, formatting and cross-checking medication interactions)
Value: It enhances patient care by reducing redundancy of procedures/tests, supporting value-based care and preventing adverse effects of medication (by built-in features notifying if any medicines prescribed can have adverse effects on each other)
Variety: It involves structured (Lab results, vitals), unstructured (physician notes, imaging reports) and semi-structured (billings) data.
-
I think Electronic health records (EHRs) combined with wearable device data from thousands of patients. for example, Apple Watch is a great example of a device that generates Big Data in healthcare.
1. Volume:
It collects a large amount of data continuously such as heart rate, steps, sleep patterns, and ECG readings.2. Velocity:
Data is recorded in real-time and can be synced instantly to the user’s phone or cloud.3. Variety:
Different types of data: numeric (heart rate), categorical (activity type), time-series (sleep patterns), and notifications from health apps.4. Veracity:
Some measurements may be inaccurate due to movement, sensor errors, or incorrect usage, so data quality varies.5. Value:
When analyzed, it helps track fitness, detect irregular heart rhythms, monitor chronic conditions, and provide insights for personal health. -
I think it is the national health records and disease surveillance data collected from public hospitals, PPM clinics and rural health centers across Myanmar. This includes patient records, diagnostic data, lab results, prescription data, COVID-19 tracking data and even mobile health data from remote areas.
Volume:
The healthcare system generates a large amount of data daily especially from hospitals in major cities like Yangon and Mandalay, but also increasingly from rural areas through digital health initiatives. This includes thousands of patient records, lab tests, and imaging data.Velocity:
During disease outbreaks (e.g. COVID-19, dengue fever), data needs to be collected and processed quickly to inform public health decisions, resource allocation, and emergency responses.Variety:
The data comes in different forms: structured (patient registration, lab results), unstructured (doctor’s notes) and semi-structured (health surveys, SMS-based symptom reporting from rural areas)Veracity:
In Myanmar, data quality can be inconsistent due to manual record-keeping, lack of standardization and underreporting, especially in rural regions. This affects trust in the data.Value:
When used effectively, this data can support disease prediction, health resource planning, and targeted health interventions especially in underdeveloped areas where resources are limited. -
-
Can you give an example of data that you think it could be considered as “Big Data”? What are the characteristics of the data that fit into 5Vs, or 7Vs, or 10Vs of Big data characteristics?
I had worked in Health System Strengthening project of Chin State in Myanamar. Considering the data I collected that time as “Big Data”, the characeristics of 5Vs are
1. Volume: State-wide health service providers, and facilities data including millions of individual and health organization records, laboratory results, vaccination data, population data.
2. Velocity: The speed is crucial to respond immediate public health threats (such as new cases of notifiable diseases, contact tracing)
3. Variety: Diverse data sources containing hospitals, HR, Equipments, commodities, and multiple registers and reports.
4. Veracity: Ensuring data quality, accuracy, and reliability is still challenging.
5. Value: When analyzed, support policymakers in strengtheing health service providing system.
-
You must be logged in to reply to this topic. Login here
