Chronic pain and general physical discomfort can be attributed to disorders or malfunctions within the Human Musculoskeletal System (HMS). To help alleviate these symptoms, Asian countries such as Japan and China have developed various traditional medicine-based exercise techniques over many years  . However, some of reviewers are critical to such non-orthodox medicine    but some others support by stating “overwhelming effort toward attempts at integrating alternative medicine into mainstream”  . In addition, these techniques have been enhanced by recent scientific research into the HMS, undertaken by biomedical and mechanical engineers.
The objective of this paper is to demonstrate the effectiveness of a practice of the clinical technique named Somatic Balance Restoration Therapy (SBRT) by applying mechanical and systems engineering methods. The SBRT is a therapist-guided self-exercise technique that helps the patient to perform easily a series of simple motions in a completely non-invasive manner. By applying a systems and mechanical engineering approach, a computerized visualization of the SBRT’s clinical technique has been demonstrated. This process was then evaluated by applying matrix algebra and correlation analysis. For the purpose of integration into conventional medicine, the terminology used by the therapist (one of the authors of this paper) is converted to the functional anatomical terms. This effort has turned out that such wording coincides with the mechanical symbols of robotic systems. Next, the evaluation was applied to a typical actual therapy records and the results revealed that the systems approach developed herein was proven valuable and gives scientific background of complementary medicine.
2. Previous Works
Chronic pain and general physical discomfort are common symptoms. However, these conditions convey important information on the clinically relevant state of the human body, especially relating to disorders or malfunctions found within the Human Musculoskeletal System (HMS)  . For many years, acute and chronical pain issues have been reported in various forms or symptoms: neck and extremity disorders among computer users, knee or hip complaints in senior adults  , and recovery after minor traffic accident injuries. Most pain research works have been conducted in Western countries such as the United States, Germany, Denmark, Sweden, France and others   . Worldwide, however, millions of people suffer from untreated pain, particularly in the developing world where the burden is highest among the poor. Reducing global inequalities in untreated pain requires a concerted effort by global health funders, institutions, and organizations. These groups must overcome the complexity of pain management by promoting multidisciplinary and holistic approaches that integrate traditional oriental techniques with Western medicine. In China and Japan, pain is widely treated by therapeutic approaches by correcting HMS distortions. Most of the above-cited studies are based on statistical analysis with a large number of samples. Lars and Swenson  conducted an overall survey on clinical findings of referred muscle pain, while Feine and Lund  assessed the efficacy of physical therapy in controlling chronic musculoskeletal disorders. In addition, a systematic review of exercise therapies that help in overcoming chronic lower back pain has been compiled. Exercise therapy has recently been recognized as part of Integrated Medicine as shown in Figure 1. However, clinical pain research is confounded by many factors that obscure specific aspects of the underlying disease, such as biases in the cognitive, emotional, and social aspects of the ailment. Moreover, pain is a multidimensional and highly individualized perception that is difficult to quantify or validate. Apart from such medical and psychological aspects of pain in the HMS, engineering approaches have been developed especially by applying mechanical engineering of the HMS methods, where anatomical motions are expressed by using mechanical element motions such as joint rotation    . However, actual motions of the real HMS are not fully related to engineering methods or tools.
In order to overcome such difficulties, this paper aims to realize a systematic interpretation of a therapist-guided technique called the Somatic Balance Restoration Therapy (SBRT)    . The methodologies used are based on mechanical and systems engineering applied to a number of therapy data recorded by one of the authors. By applying this approach the SBRT emerges as an academically appropriate technique representing a step forward in exercise therapy. With clear visualization of the method, this technique can be used as an educational tool for the inexperienced practitioner and as a management tool for established practitioners.
3.1. Therapist-Guided Exercise: Somatic Balance Restoration Therapy
Figure 2 shows a flow diagram of the SBRT system approach which begins with a “Therapist-Guided Motion Test”. During this test, the patient lays in a face up (supine) or face down (prone) position and performs a series of guided body movements called Active Motions shown in Figure 3 and Table 1. There are basically 40 active motions with left and right directions, that makes 80 motions altogether. The therapist systematically guides the patient to try each of the active motions from #1 to #80. It is important to tell the patient to perform it only if it is easy, and not to perform the motion if the patience feels pain or hard to do so. In short, the therapist will attempt to guide the patient through each Active
Figure 1. Integrated medicine scheme.
Figure 2. Flow diagram of the SBRT process.
Figure 3. Eighty active motions of SBRT.
Motion (patient initiated motion) listed in Table 1 and to record the results in the SBRT sheet. It is noted that Motions from #81 to #138 in Table 2 (to be called Associate Motions and not performed intentionally by the patient) are
Table 1. List of active motion.
Table 2. List of associated motion.
induced only by the Active Motions from #1 to #80 in Table 1. Thus, the whole Active and Associated Motions are called Fundamental Motion Elements (FME).
The goal of the SBRT is to eliminate or at least alleviate overall pain or discomfort, which includes any pain during the therapy. This is achieved by the patient reporting varying degrees of difficulty while performing the motion test. As mentioned, the therapist does not have physical contact with the patient during the examination or the guided self-exercise phase of the therapy. However, to assist the patient’s performance of certain motions, the therapist may apply slight resistance against the patients’ limbs while performing problematic motions.
3.2. Relation between Active Motions and Anatomical
A patient can perform any of the Active Motions with intension to do it, but usually cannot perform a single anatomical motion.
For example, if the patient intends to extend arm to left (Active Motions #5), it is realized by horizontal adduction and retraction of the shoulder together with flexion of the elbow and extension of the wrist. It is noted that the anatomical motions correspond to the joint degrees-of-freedom of 15 body human model of Figure 4 and its robotic symbol expression of Figure 5. Table 3 shows the relation between active motions and anatomical motions. Such relations are conveniently represented by matrix format to be developed in the following section.
3.3. Matrix Representation of Interconnecting Motions
Among the 138 FMEs, the 80 Active Motions are intentionally performed by the patient (Table 1); the remaining 58 are Associated Motions induced by Active Motions (Table 2). FMEs are numbered #1 to #80 for Active Motions (Table 1). Of the Associated Motions in Table 2, FMEs #81-#109 are induced while in the face up (supine) position, while FMEs #110-138 are induced while in the face down (prone) position. Please note that although FMEs #1-80 are typically Active Motions, they can also act as Associated Motions in certain instances.
Figure 4. 15 body model.
Figure 5. Motion diagram represented by JIS B0138.
Since the HMS is interconnected in a highly complex manner, each individual segment or component of it requires an integrative approach    . Active Motions are also interconnected with other Active and Associated Motions. Therefore, the relationship between any two of the 138 FME can be expressed by an N-square matrix  of dimension 138. Likewise the relationship between Active Motions and Fundamental Motion Elements is represented by a 80 by 138 matrix to be called Matrix-A.
The above-mentioned processes are interpreted in terms of orthodox medicine as shown in Table 3. It is noted that an Active Motion induces one or more joint motions in most case, and human body motion can be realized in a combination of a few joint motions. Table 4 will be useful to bridge the gap between orthodox and practitioner medical approaches for dealing musculoskeletal aspects of the HMS. In order words, the 138 FMEs is related to some of the 80 joint motions that correspond to anatomical motions, and this relation is represented by a matrix of dimension 138 by 80 to be called Matrix-B.
4. A Systems Approach to Identify Malfunctions and to Select Remedying Motions
The SBRT process is digitized and visualized by extensive use of matrix representation, where the elements of Matrix A and B are denoted by
Table 3. Articulation names of joints in anatomical posture.
Note: Only left limb joints are cited.
Table 4. An example of SBRT record (partially shown).
Note: R/L = Right/Left, h = hard, e = easy, na = not applied, B-2, F-1, F-5, G-8 etc: Body portions (See Figure 3).
and respectively, then we can then we can define diagnostic matrices, Q by
The Active Motion during the Motion Test is weighted by quantity wi, whose numerical value reflects the results (easy, hard or painful, denoted e, h and p, respectively). The weighting values are determined subjectively yet skillfully by the therapist or the analyzer through experience. Note that easy movements are assigned large positive weightings, while painful movements are negatively weighted. We evaluate the selection of the Active Motion by computing correlation coefficients of the joint diagnosis matrix. The correlation coefficient is given by
where, in the joint diagnosis matrix of Equation (1),
Traditionally, the SBRT therapist (and other Eastern medicine therapists) directly records patient examination results into a simple human-body model, and then identifies the malfunctioning point as the remedial exercise motion. Because expert therapists are relied on many years of personal experience and success to diagnose and treat patients, this knowledge was not wholly available to successors. This difficulty in transferring skills and experience to the next generation has motivated us to development a computer software support system that allows successors to learn from expert therapists and their experiences. The process of this therapeutic approach, as applied to actual cases is detailed below:
1) From the motion test results, identify the body motions that cause pain or discomfort
2) List the Active Motions associated with the pain-inflicting motion(s)
3) Identify the joint movements and muscle groups that are associated with the painful motions
4) Search for a frequency distribution pattern in the identified joint movements and muscle groups
5) Identify the joints and muscles associated with the highest frequency or reoccurrence of reported pain as the most probable causes of the malfunction(s)
6) From the identified rotary joints and muscle groups associated with the highest frequency of pain reports, find the Associated Motions using the charts of relationships between FMEs and Joint Motions
7) Identify the Active Motion that is interconnected with the Associated Motion, limiting the search to motions that can be comfortably performed by the patient
8) Identify the frequency distribution of the identified Active Motion
9) The Active Motion most frequently reported as comfortable will be applied as the remedial exercise motion in therapy
Note that Steps 1 - 5 identify candidate locations of the malfunction, while Steps 6 - 9 identify the corrective motions for therapy.
The Flow Diagram of Figure 2 is a visual representation of the computerized SBRT process. The inputs are therapeutic results constructed from the list of the pain-causative motions listed in Table 4. From these, the relationship(s) between pain-causative Active Motions and joint DOF are generated. The diagnostic matrix Q is then determined by the identification of these relationships.
The results of the patient feeling, easy, hard or painful, are recorded on the SBRT sheet as illustrated by an example shown in Table 4. In this example, the patience can perform Motion #41 (turning neck to right) easily, but cannot do Motion #1(turn neck to left), because the patient feels pain at a specific location B-2 in Figure 6 and Figure 7.
Through the proposed process we can identify comfortable motions that induce joint motions consequent to Associated Motions. The goal of the SBRT is to find comfortable Active Motions that activate Joint Motions reported as painful. However, in general, these motions are not uniquely specified and thus the therapist must select the most effective motions among multiple comfortable Active Motions. For this purpose, we apply the cross-correlation analysis together with the “Distance Measure”, defined as the number of joints between the malfunctioning zone and the motion-related joint.
The objective is to select or specify the Active Motion(s) that most effectively
Figure 6. Partitioning of human body.
activate the malfunctioning area as identified in the previous section. For this purpose, we evaluate the painful and easily performed Active Motions through cross-correlation analysis typified by the Correlation Coefficient evaluation defined by Equation (3). Figure 8 shows the correlation coefficients between 21 “easy” Active Motions and 4 “painful” Active Motions, numbers 1, 2, 4 and 6. From Figure 8, we can easily identify Active Motions candidates that are suitable for remedying HMS parts causing pains.
We observe that there are still two or more candidates to remedy the pain, and in such cases we will apply other selection criteria by defining another criterion of optimization.
5. Concluding Remarks
The intention of this paper is to bridge the gap between conventional medicine and traditional therapy by removing obstacles and applying multidisciplinary approaches based on the systems engineering, and this has been attained to some degree. To demonstrate such an integrated approach, we have taken a therapist-guided exercise for restoring human musculoskeletal balance called the Somatic Balance Restoration Therapy (SBRT). With this approach the practitioner has converted terminology used in therapy into wording of functional anatomical terms, and this effort has turned out to be useful for communication between the specialists of three different disciplines, namely, therapy, conventional medicine and systems engineering. Some examples have demonstrated a first step of
Figure 7. Seven active motions with pain in Table 3.
Figure 8. Correlation between active and painful motions shown in Table 4.
integrated and systematic approach for identifying malfunctions and remedying corrective exercise within the Fundamental Motion Elements.
The authors wish to express their appreciation to Dr. Motonaga of the Miyako-Jima Prefecture Hospital for his valuable advice.
FME Fundamental Motion Elements.
HMS Human Musculoskeletal System.
SBRT Somatic Balance Restoration Therapy.