The Design and Evaluation of an Intelligent Pain Management System (IPMS) in Cancer Patient Care

The Design and Evaluation of an Intelligent Pain Management System (IPMS) in Cancer Patient Care

Y. Ken Wang, Juan J. Gu, Yunheng Sun, Feng Jiang, Hongwei Hua, Jing Li, Zhijun Cheng, Zhijun Liao, Qian Huang, Weiwei Hu, Gang Ding
Copyright: © 2020 |Pages: 20
DOI: 10.4018/978-1-7998-0047-7.ch011
(Individual Chapters)
No Current Special Offers


This case study reviews the design and development of a mobile-based intelligent pain management system (IPMS) app in cancer patient care and pain management in a rural hospital in China. Healthcare professionals were involved throughout the design to the evaluation stages. The IPMS facilitated real-time pain recording and timely intervention among cancer patients with pain. To evaluate the effectiveness of the IPMS, a clinical trial was administrated under the supervision of healthcare professionals. The result confirmed that the IPMS was a feasible, effective, and low-cost pain management tool for cancer patients and healthcare professionals. This case provides preliminary data to support the potentials of using IPMS in cancer pain management and emphasized that the involvement of healthcare professional throughout the system development lifecycle is crucial to the successful implementation of the IPMS.
Chapter Preview


Cancer represents a group of diseases characterized by uncontrolled growth and spreading of neoplasm cells in part of the body (Mathur, Nain, & Sharma, 2015). The World Health Organization (WHO) estimated that 14 million new cancer patients were diagnosed worldwide every year and cancer resulted in 8.2 million deaths in 2012 (World Health Organization, 2017). As a leading cause of human death globally, cancer, as well as cancer treatment became a focus of medical research for many decades.

Typical symptoms of cancers include weight loss, unexplained bleeding, and pain (Linder et al., 2015). Cancer patients, especially at their terminal stages, often experienced cancer pain. Prior research found that over one-third of cancer patients experienced cancer pain (Deandrea, Montanari, Moja, & Apolone, 2008). Cancer pain has been considered a major reason that caused a lower quality of life of the patients (Lesage & Portenoy, 1999; Portenoy & Lesage, 1999). Under-treatment of cancer pain has been a worldwide problem (Deandrea et al., 2008; Greco et al., 2014). Effective management and further mitigation of cancer pain requires an accurate and precise assessment of the pain (Forbes, 2011; Jacobsen et al., 2009; Kuzeyli Yildirim & Uyar, 2006; Kwon, 2014).

Conventional paper-based, self-reported pain questionnaires are limited in terms of efficiency and accuracy. Ward et al. (1993) surveyed 270 patients with cancer and found that patients were reluctant to share their pain information in self-reported questionnaires due to privacy and other concerns. Paper-based pain reporting methods are prone to potential biases in the collected data due to patients’ lack of relevant knowledge. Ward et al. (2014) reported that elderly, less educated, and low-income patients are more likely to have concerns at pain reporting and the levels of concerns were correlated to the levels of pain. Particularly, under-medicated patients had significantly higher concerns to report their pain possibly due to anxiety. Such limitations call for a more efficient and accurate method in pain reporting and management. Modern technology, such as visual interactivity and touch screens may help those patients better understand the reporting procedures and provide effective guidance for them to report pain.

Rapidly growing mobile technology use gave rise to the emergence of more advanced electronic pain reporting and assessment systems (Agboola, Ju, Elfiky, Kvedar, & Jethwani, 2015; Heinonen, Luoto, Lindfors, & Nygard, 2012; Jan et al., 2007; Marceau, Link, Jamison, & Carolan, 2007; Mulvaney, Anders, Smith, Pittel, & Johnson, 2012; O'Reilly & Spruijt-Metz, 2013; Stone & Broderick, 2007). Such systems, often in the form of smartphone applications (apps), sometimes referred to as intelligent pain management systems (IPMS), can be more effectively capturing, transferring, and analyzing the pain data (Sun et al., 2017). Marceau et al. (2007) introduced PDA (personal digital assistant) devices to pain management. Over the two weeks experimental period, chronic pain patients were asked to monitor their pain, mood, activity interference, medication use, and pain location on paper or on PDA devices. The participants reported that the use of PDA devices was more convenient, time-saving, and easier to report data.

Key Terms in this Chapter

Life Quality: Life quality (also known as quality of life) is an overall assessment of the important aspects of patients’ life, including the status of patients’ cognitive, emotional, physical, social, and spiritual aspects as well as the patients’ health care and personal autonomy aspects.

Karnofsky Performance Status (KPS): The Karnofsky Performance Score (KPS) was developed by Dr. David A. Karnofsky and his colleagues to evaluate patients’ health status. The KPS ranking describes patient’s health status from 0 to 100, with 0 being patient death to 100 being patient in perfect health condition. In practice, the KPS scores are often evaluated in an interval of 10.

Intelligent Pain Management Systems (IPMS): An intelligent pain management system (IPMS) is a computer system that can effectively capture, transfer, store, and analyze patients’ pain data with minimum intervention from the healthcare providers. A recent IPMS often includes a module to support mobile devices. It is expected that future IPMS will incorporate new advancements from the research of artificial intelligence, cloud computing, and the fifth generation of mobile communication system (5G).

Mobile Applications: Mobile applications (also known as mobile apps) are software programs that operate on mobile devices, including smartphones, tablets, wearable devices, etc. Mobile apps are often optimized for mobile devices to offer location-based, personalized, and time-sensitive features. Examples of mobile apps include social networking, maps and navigation, communication, shopping, personalized education, etc.

Clinical Trial: Clinical trials are experiments conducted in clinical research and practice to discover the effect of particular treatments or interventions. Clinical trials are usually conducted with at least an intervention group and a control group, in order to identify the true effect without interference from uncontrolled variables.

Cancer Pain: Cancer pains are usually caused by tumors pressing on bones, nerves or other organs in the body or cancer treatment such as chemotherapy or radiotherapy. Cancer pains include nerve pains, bone pains, soft tissue pains, phantom pains, and referred pains. Chronic cancer pain, often due to nerve changes, may cause mild to severe pain for a long period of time. Controlling cancer pain is a critical part of cancer patient care, especially for patients at the terminal stages.

Numerical Rating Scale: The Numeric Rating Scale is a quantitative measure of patients’ pain intensity. In practice, patients are often surveyed with questions related to their perceived level of pain. A patient selects a whole number (from 1 to 10) that best reflects the intensity of his or her pain.

Pain Management: Pain management employs a combination of medical, physical, and psychological approaches to ease the painful feeing of patients and to improve their quality of life. Pain management methods involve medication, psychological counseling, physical therapies, nursing care, etc. Effective management of chronic pain is important for cancer patients.

Machine Learning: Machine learning is a technology that relies on computer algorithms and statistical methods to automatically generate patterned models from sample data sets without human intervention. The sample data sets, also known as the training data , may be labeled (containing both inputs and known outputs) or unlabeled (only containing inputs without outputs). If a patterned model is generated from labeled data sets, the process is called supervised learning. If a patterned model is generated from unlabeled data sets, the process is called unsupervised learning.

Complete Chapter List

Search this Book: