Application of Health Information Technology in Health Informatics

Health information is data about a person’s medical history, such as symptoms, diagnoses, procedures, and outcomes. Patient histories, lab results, x-rays, clinical information, and notes are examples of health information records. A patient’s health information can be viewed individually to see how their health has changed; it can also be viewed as part of a larger data set to understand how a population’s health has changed, and how medical interventions can change health outcomes. In the application of health information, there is a management system in place to practise the acquisition, analysis, and protection of digital and traditional medical information viral to provide quality patient care. It is a synthesis of business, science, and information technology.

Professionals in health information management work in a variety of settings and job titles. HIM functions as a bridge between clinical, operational, and administrative functions. This professional has an impact on the quality of patient information and care at every point in the healthcare delivery cycle. HIM professionals care for patients by managing their medical data. HIM professionals are accountable for the quality, integrity, and security of patient health information. Based on health information, there is also health information technology HIT, which refers to the framework used to manage health information and the exchange of health information in a digital format. HIT focuses on the technical aspects of managing health information, such as working with software and hardware used to manage and store patient data. Health informatics (HI) is a science that defines what is technically captured, transmitted, and used. Health informatics is concerned with information systems, principles, and technology as they apply to the continuous management of healthcare delivery.

EXAMPLE IN HEALTH INFORMATICS

Deep Learning For Disease Diagnosis

Diagnostics is one of the areas of artificial intelligence in health care that has shown the most promise. Early diagnosis is one of the most important factors in determining the ultimate outcome of a patient’s care. Deep-learning AI algorithms are being used to shorten the time it takes to diagnose serious illnesses. The way artificial intelligence rapidly processes large amounts of information and arrives at likely causes of symptoms can drastically shorten the diagnosis-treatment-recovery cycle for many patients. The consequences of this are already being felt in a variety of areas.

The authors describe a machine learning tool developed at the University of Virginia School of Medicine that can rapidly analyse thousands of images from children’s biopsies and distinguish between environmental enteropathy and celiac disease in the article “Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children,” which was published in the Journal of the American Medical Association. It is hoped that this will expedite children’s treatment and reduce rates of stunted growth that can occur as a result of delayed diagnosis.

Deep learning for medicine is a relatively new development that has yet to be thoroughly investigated. To estimate the performance of deep learning algorithms in health care, a search was conducted across several databases using the following search terms: (‘deep learning’ OR ‘neural network’ OR ‘machine learning’) and I medical imaging (ii) EHR (iii) genomics (iv) sensing and online communication health. Among the articles discovered, papers that were significantly relevant to each part and used DL algorithms were briefly reviewed.

Medical Imaging

The first applications of deep learning on medical datasets were medical images such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET), X-ray, Microscopy, Ultrasound (US), Mammography (MG), Hematoxylin & Eosin Histology Images (H&E), Optical Images, and so on. PET scans reveal regional metabolic information via positron emission, as opposed to CT and MRI scans, which reveal structural information of organs or lesions within the body in perspective with radio waves, X-rays, and magnets.

Medical imaging technology was chosen for purposes, and in terms of potential health risks to the human body from X-rays, low-dose CT scans were also considered, but have drawbacks such as image quality and diagnostic performance. Pathology, psychiatry, brain, lungs, abdomen, heart, breasts, and other applications have been studied, as well as image classification (classify disease present/absent), object detection (detect disease with location), image segmentation (detect disease and label pixels), image registration (transform one image set into another set of coordinate systems), and other tasks. Image classification remains the preferred method for medical image research, with each image classified into one or more classes. Its limitations include a lack of labelled training samples, which has been addressed by transfer learning and multi-stream learning.

A combination of RNN and CNN was also investigated in order to track disease progression and fully utilise 3D data. Deep learning has also been quickly implemented in all other aspects of medical image analysis, such as pixel, edge, and region-based image segmentation, class imbalance studies, image registration (for example, registration of brain CT/MRI images or whole-body PET/CT images for tumour localization), image generation, image reconstruction, and so on.

Machine Learning and Radiology

Machine learning is a statistical technique for fitting models to data and learning by training models with data. In a 2018 Deloitte survey of 1,100 US managers whose organisations were already pursuing AI, 63 percent of companies surveyed were employing machine learning in their business. It is a broad technique at the heart of many approaches to artificial intelligence, and there are numerous variations on it.

Precision medicine is the most common application of traditional machine learning in healthcare, predicting which treatment protocols are likely to succeed on a patient based on various patient attributes and the treatment context. The vast majority of machine learning and precision medicine applications require a training dataset for which the outcome variable, such as disease onset, is known; this is known as supervised learning.

A more complex form of machine learning is the neural network – a technology that has been available since the 1960s and has been well established in healthcare research for several decades and has been used for categorisation applications such as determining whether a patient will acquire a particular disease. It considers problems in terms of inputs, outputs, and variable weights or ‘features’ that link inputs and outputs. It has been compared to how neurons process signals, but the analogy to brain function is relatively weak.

Deep learning, or neural network models with many levels of features or variables that predict outcomes, is one of the most complex types of machine learning. There could be thousands of hidden features in such models, which are revealed by the faster processing of today’s graphics processing units and cloud architectures. Recognition of potentially cancerous lesions in radiology images is a common application of deep learning in healthcare. Deep learning is increasingly being applied to radiomics, or the detection of clinically relevant features in imaging data that go beyond what the human eye can see. Both radiomics and deep learning are commonly used in oncology-related image analysis. Their combination appears to promise greater diagnostic accuracy than the previous generation of automated image analysis tools, known as computer-aided detection or CAD.

Deep learning is also increasingly used for speech recognition and, as such, is a type of natural language processing (NLP), which is discussed further below. Unlike earlier forms of statistical analysis, each feature in a deep learning model typically has little meaning to a human observer. As a result, explaining the model’s outcomes may be difficult or impossible.

When clinicians must distinguish between distinct but related conditions, interpreting clinical biopsy images for disease diagnosis can be difficult. Recently, there has been a surge of interest in artificial intelligence methods that assist clinicians in converting big data (biomedical images and patient bio sample data) into accurate and quantitative diagnostics.

To our knowledge, the majority of computer modelling advancements in health care, particularly in image analysis, have focused on feature engineering, i.e., asking a computer to evaluate pre-specified, explicit image features to allow computational algorithms to detect disease or specified lesions. Deep learning, also known as a convolutional neural network (CNN), is a type of artificial intelligence that employs machine learning techniques to process and interpret data (by detecting and segmenting multiple pixel intensities within a single image and labelling features at a pixel-by-pixel level).

Administrative Task Automation

Automation is the reduction of manual tasks through the use of information technology or Artificial Intelligence (AI). Automation is most beneficial to repetitive tasks that require little human intervention. Data automation is the process by which programming handles data assimilation, storage, and analysis.
Automation in healthcare is one of the most significant benefits of Artificial Intelligence (AI). We frequently overlook these facilities. AI does not always have to be about the most eye-catching aspects of technology, such as robots in surgery. Using AI automation in basic operations and administration tasks can improve patient experience, service quality, project implementation, and reduce costs.

As a result of the introduction of this new method, the proper infrastructure is required to support high performance computing systems. The criteria should be quick, dependable, and capable of handling large amounts of data. Aside from radiology, automation can have a significant impact on many operational and administrative aspects of the healthcare system. We can easily automate time-consuming tasks with good strategic thinking and planning. Examples of applications of administrative task automation in the healthcare sector are provided below.

Data Automation (data collection from patients)

The use of automated processes, equipment, or systems for the purpose of collecting, processing, and storing data obtained from a source, such as optical scanners, bar code readers, or magnetic stripe readers, is referred to as data automation. Big Data Automation, for example (BDA). It is a broad concept that creates value through the use of innovation and IT resources.
Typically, data collection automation entails extraction, transformation, and loading. Extraction is a collection of open data sources for relevant information. Data transformation is the conversion of data into a machine-readable format. Loading refers to the data that must be fed into the system in order for it to serve as the raw material for automation.
Automation data systems can help to reduce manual labour work in data collection and analysis; additionally, this automation system can significantly reduce administrative workloads. It has the potential to improve the patient care system in a variety of ways.
Aside from that, instead of taking notes on client feedback and complaints over the phone or sorting through emails, online forms can help to collect data about the customer’s information, need, enquiry, and complaint. Instead of relying on personal research, online forms can aid in data collection. There is no need to manually scan or enter anything. A form will do the grunt work for you.
We can include questions about what our customers are concerned about or looking for in those electronic forms on the company’s website. The information provided by the customer can assist you in better understanding and meeting their needs. Nowadays, there are many automation systems that can go a step further and categorise customers based on information. The company can then devise a series of follow-up actions to entice them to become leads.

Sorting and responding to emails can be automated.

Email is a common method used in many industries, including healthcare, to communicate, send information, inquire, and receive information. It resulted in an overabundance of email traffic in the inbox. Employees spent a lot of time checking email at times. To reduce the amount of time employees spend checking email, a label, such as a smart label in Gmail, that automatically sorts emails as they arrive, should be set up. As a result, we won’t have to spend as much time going through each email one by one. To avoid responding repeatedly, email templates and canned responses can be created.

Making appointments

Scheduling meetings with multiple people would be difficult, especially when everyone is as busy as you are. It is difficult to find a time that works for everyone’s schedule. It will be much easier to use software that can sync everyone’s calendars, such as Google Calendar or iCal. Aside from meetings, this calendar can be used for a variety of purposes. Make a time block to complete the most important tasks first.
The cost and time savings in scheduling patients’ appointments and automated reminders can help free up valuable staff, allowing them to focus on patient care.

Bill payment

Investing in an automated data system that can manage all payments in one location can assist in avoiding the need to cut checks every month and manually enter payments into accounting systems. Many businesses have implemented an auto-payment system. This way, we never have to worry about missing a bill payment. The payment will be set up in advance so that the funds are automatically deducted from the account.
Answering customer questions
People nowadays are dissatisfied if a service is not provided in the time frame that they expected. As a result, the customer service team is frequently faced with a challenge. The customer will be impatient. They anticipate a prompt response and feedback. Many businesses are now changing their settings to automation in order to alleviate this problem. We can now configure automated email responses so that when a customer sends a query, the automated email system responds.
In some businesses, a help desk system is set up within the customer service team. Instead of allowing all customer emails to accumulate in an inbox, the system can sort and route the emails based on criteria the customer specifies, ensuring that the emails reach the right person and are answered as soon as possible.

Cut Operational Costs

Health informatics encompasses a wide range of approaches to improving healthcare services, such as changing the way healthcare is funded to increasing efficiency in the United States. According to a new study published by global consulting firm PricewaterhouseCoopers (PwC), medical informatics technology is critical in improving the general population’s health and lowering the costs of providing quality healthcare.
The survey included more than 600 health management professionals, and they discovered that 79 percent of respondents want clinical informatics technology to help them eliminate medical errors, and 52 percent want electronic health records to help them reduce operational costs. In addition to lowering costs and educating patients about their own healthcare, many facilities hope to use medical informatics systems to reduce hospital readmission rates, particularly those related to complex health conditions such as cancer.

Nonetheless, one area identified by the study as currently lacking, but with potentially significant benefits, is the use of healthcare IT in educating patients on healthy lifestyle management. According to the report’s findings, only 15% of health insurers and 13% of hospitals have been able to leverage the power of technology to educate patients, despite 61 percent of respondents saying this was a primary goal of their healthcare IT approach.

Nearly 84 percent of participants said that standardising medical data remained a significant challenge for them, and that healthcare IT vendors, care providers, and pharmaceutical companies needed to collaborate to develop industry standard information formats. Jerry Buchanan, account director of healthcare technology and services at eMids Technologies, discusses five ways that health informatics can help to reduce healthcare costs in the long run.

1. Higher levels of care
   It will provide the best treatment methods and contribute to a healthier population by analysing data collected by electronic health records. “Whether or not this data is combined with financial data to analyse cost effectiveness… is tangential to the overall goal of determining the best way to handle treatment for each individual patient,” Buchanan explained.
2. Increased patient participation and collaboration
   According to Buchanan, health IT “provides a clear avenue for enterprising organisations to develop innovative disease management solutions to address the issue.” The information gathered would be extremely useful in determining ways to reduce the costs associated with chronic illness.
3. Putting information in the spotlight
   Because the healthcare industry is constantly evolving, there is an overwhelming amount of information to distil and absorb. Health IT provides a means to disseminate that information.
4. Concentrate on outcomes
   “The oncoming tidal wave of electronic clinical data presents an opportunity to replace our outmoded, volume-based, fee-for-service business model with one focused on product quality,” Buchanan said.
5. Openness to the patient
   Health IT must be used as a mechanism to involve patients in their own healthcare. “Our current healthcare financing system completely shields patients from the cost of their care,” said Buchanan. “Until we find a way for patients to educate themselves and question services, quality, and price, market forces that can naturally contain rising healthcare costs will never be able to work.”

CONCLUSION

Information is a vaporous commodity. According to one definition, it is the data and knowledge that intelligent systems use to make decisions. Health informatics assists doctors in their decisions and actions, and improves patient outcomes by making better use of information to capture, process, communicate, and apply patient data and medical knowledge in a more efficient manner.
These challenges have grown in importance as the internet has made it easier for patients to access medical information. There are numerous advantages to using health informatics. For example, by using a computer, the internet, and various medical databases, doctors can learn how to treat patients more effectively. Additionally, the data retrieved by using health informatics provides statistical information that can be used to make decisions.

REFERENCES
Deloitte Insights State of AI in the enterprise. Deloitte, 2018. www2.deloitte.com/content/dam/insights/us/articles/4780_State-of-AI-in-the-enterprise/AICognitiveSurvey2018_Infographic.pdf.
Journal of the American Medical Association, “Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children”
UIC, AI in Healthcare: 4 Examples in Health Informatics. https://healthinformatics.uic.edu/blog/ai-in-healthcare-4-examples-in-health-informatics/
Study.com. 2020. [online] Available at: https://study.com/academy/lesson/benefits-functions-of-health-informatics.html [Accessed 16 April 2020].
Healthitoutcomes.com. 2020. Health Informatics and The Future of Healthcare. [online] Available at: https://www.healthitoutcomes.com/doc/health-informatics-the-future-of-healthcare-0001 [Accessed 16 April 2020].
Ahima.org. 2020. HIM Careers – Health Information 101. [online] Available at: https://www.ahima.org/careers/healthinfo [Accessed 20 April 2020].
Chandra, R. (2016). 28 Murid SD diduga Keracunan. Padang Ekspres, 19–30. https://doi.org/10.1016/j.jhealeco.2013.12.005.The
Now, A., & Details, G. P. (2020). Medical informatics vital for continued cost reductions and better health, says report. 1–2.
Kilbrink, N., & Bellstrom, P. (2010). Problem-Based Learning in a Programming Context–Planning and Executing a Pilot Survey on Database Access in a Programming Language. Information Systems Development Towards a Service Provision Society, 867–875. https://doi.org/10.1007/b137171
Sujan Patel, 6 Administrative Tasks You Can Automate (Are you being productive or busy)?,2020,https://www.inc.com/sujan-patel/6-administrative-tasks-you-can-automate.htmlJosh Gluck, How Automation in Healthcare is Boosting the Bottom Line, 2019, https://healthtechmagazine.net/article/2018/06/how-automation-can-translate-better-patient-care-and-boost-bottom-line