3.5.3 Patients Predictions for Improved Staffing
Main big data applications in the medical industry are the patient’s predictions. The prediction of anticipated number of patient’s visits to emergency will improve the staffing and provide a way to utilize the resources of the hospital properly. One of the major problems in hospitals is staffing, that is how many people (doctor, nurse, and other workers) are required at any given time period. Big data is used to solve this problem.
Intel and the Assistance Publique-Hôpitaux de Paris (AP-HP) worked toward developing a cloud based system for predicting the anticipated amount of patient’s visits [3]. The overcrowding is an emerging problem in most of the hospitals. The cloud based big data analytics system will predict the patient visits and the admissions of patients in hospitals as inpatient or in emergency ward. These solutions make it simple for the hospital administration to productively use the resources (staff, beds, etc.) of the hospitals. The advantages of patient’s prediction are improved patient services, reduced wait time, and reduced labor costs, and this will lead to better care with lower cost for patients.
3.5.4 Medical Imaging
Big data analytics for healthcare read the images using algorithms. These algorithms analyze the image by identifying patterns in the pixels and convert it into numbers. The result of the analysis is used by the doctor during the diagnosing process. Because of big data analytical tools, the doctor will analyze the outcomes of the algorithms.
Artificial intelligence is more efficient for the diagnosis of many diseases. One of the examples of AI is diagnosing breast cancer, which is developed by scientists of Google. This diagnosing system uses deep learning algorithm to predict the cancer cells which uses medical images as input [15]. Artificial Intelligence with medical imaging is used in early diagnosis of neurological disease, cardiovascular abnormalities, and cancer cells [30]. Medical imaging with the big data analytical tool and artificial Intelligence is mainly used in the treatment and disease diagnosis process in the early state.
The other big data applications in healthcare industries are as follows:
• Automation of hospital administration process
• Personalized healthcare service
• Medical research and the drug discovery
• Prevent health insurance fraud
• Prevent unnecessary emergency visits of patients
• Practice telemedicine
• Enhance the data security
• Improved strategic planning with the health data.
All the abovementioned applications of healthcare, in big data, can save human lives with more accurate and personalized patient care. The applications of healthcare in big data have a great impact on our lives. The researchers, hospitals, and the physicians are continuously analyzing the vast amount of healthcare data for developing new therapies and for finding new treatments. The innovation in healthcare application is evolving due to continuous development in technologies. The big data application in medical industries helps to prevent pandemic and disease with lower cost.
3.6 Challenges with Healthcare Big Data
3.6.1 Challenges of Big Data
The main challenge in big data is to handle and manage the huge amount of assorted, complex, and interconnected information. It is difficult to sort out the data and prioritize the data because of its volume and variety.
Major big data challenges are as follows:
• Lack of proper understanding of the big data
• Data growth
• Combining data from various sources
• Selection of big data tool to capture, store, process, and analysis
• Data security
3.6.2 Challenges of Healthcare Big Data
The healthcare data and its analytics have problems like data accuracy, data security, data sharing, data capturing, and data visualization. The following are the major challenges faced by healthcare organizations with big data [10, 21, 22, 28].
Data Capturing
The main problems in healthcare big data are capturing data. The qualities of the clinical decisions are based on how well the big data is captured. The healthcare organization improves their data capture method by prioritizing the data based on the values.
Data Cleaning
Data cleaning is essential as medical data is captured at a variety of data sources. The data cleaning process checks the data for accuracy and correctness and also ensures whether the data are relevant.
Data Storage
The volume of healthcare data grows exponentially with time. It is difficult for the healthcare organization to store the big healthcare data with the traditional database and in a single system.
Data Security
Data security is the major challenge in the healthcare industry because of the frequent hackings. The challenges of healthcare providers are data encryption, data masking, and other protection methods for the limited data access to the outside people.
Lack of Integration between Administrative Data and Clinical System
Sometimes, there is a gap between the administrative data and the clinical system. Example: variation in the treatment code and the care given to the patients.
Data Sharing
The standard of collecting and storing the big data is different from one healthcare organization with another healthcare organization. The sharing and integrating the different formats of medical records are the real problem with medical data.
Data Updating
The healthcare data is not fixed. Most of the patient data requires frequent updates. For better treatment the up to date and real-time patients’ data is necessary. For example, the BP of the patient will vary every day. The old data will lead to wrong decisions in patient’s healthcare practice.
The main challenge in healthcare data is maintaining balance between protections of patient’s sensitive information, data integrity, and the usage of data for the knowledge discovery process [23]. All these challenges of healthcare data are not insoluble. The advanced technologies and new innovations will address all the abovementioned challenges of big data.
3.7 Conclusion
The objective of this chapter is to give a general idea about overview and characteristics of big data, different steps for deriving value from big data, big data technologies used for every step in the value chain process, use of knowledge systems in big data healthcare, various big data applications, and big data challenges in healthcare organizations. The healthcare organization needs to devote time and resources for implementing the big data value chain. Health is the precious gift of God to humans. A personalized and sustainable patient care service is the need of the hour. Big data analytics is capable of providing knowledge and valuable insights in medical data, which are useful for clinical decision, disease prediction, better treatment, etc. Healthcare big data is a permanent and continuously increasing phenomenon, which needs newer tools and technologies