|
 
 |
| COMMENTARY |
|
| Year : 2017 | Volume
: 3
| Issue : 4 | Page : 150-153 |
|
The increasing role of use of computer-based methods in disease diagnosis
Thanapong Chaichana1, Zhonghua Sun2, Anthony Lucey3, David Reid1, Atulya Nagar1
1 Department of Mathematics and Computer Science, Liverpool Hope University, Liverpool, England, UK
2 Department of Medical Radiation Sciences, School of Science, Curtin University, Perth, Australia
3 Department of Mechanical Engineering, Curtin University, Perth, Australia
| Date of Web Publication | 26-Mar-2018 |
Correspondence Address:
Thanapong Chaichana
Liverpool Hope University, Liverpool L16 9JD
UK
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/digm.digm_40_17
How to cite this article:
Chaichana T, Sun Z, Lucey A, Reid D, Nagar A. The increasing role of use of computer-based methods in disease diagnosis. Digit Med 2017;3:150-3 |
In the modern world, with its advanced health care, it is usual, and to a large degree expected, that people live longer and healthier lives. However, most modern health-care organizations still rely on practices and functions that date back to their origins in the middle of the last century. Since then, technology has made significant advances, resulting in enormous progress with more accurate medical diagnosis, better treatment, and efficient health-care delivery. However, due to various pressures on health-care organizations and legacy clinical practices, these advances are not being fully utilized as quickly as they should.[1] In the past, patients were often treated as passive recipients of treatment. Now, it is increasingly expected that patients play a more active role in understanding and choosing the options of their treatment. More emphasis is given to the patients' understanding of their disease and condition rather than just supporting them and managing their illness. An unintended problem with these advances is that the vast quantity of information generated by advances in computing, medical smart technologies, and medical software can exceed the comprehension of the average patient. Thus, there is a growing tension between the volume of data that advanced technology generates, patient autonomy, and the ability of the patient to understand what is, and what is not, a sensible course of treatment.[1]
All computerized systems employed in medicine are concerned with data; this could be discovery, management, translation, or visualization of data, or be a combination of all or several of these functions. However, it is in the visualization of the data, and the way it is chosen to be presented to the patient, that will have the biggest influence on a patient's understanding of an illness or proposed course of treatment. Understanding medical images and medical information is a crucial prerequisite for developing a patient's treatment plan.
The collective activities of medical modeling, large-scale computation, simulation, and the use of smart technologies is often referred to as “digital medicine.” The overt aim of digital medicine is to gain the new knowledge and to develop noninvasive techniques for the diagnosis and assessment of diseases in patients. The current state of the art in noninvasive imaging uses machine learning in medicine and health care in order to better inform decision-making and, ultimately, the patients themselves.
In this commentary, examples of noninvasive imaging techniques that support medical diagnosis and education are overviewed. A schematic diagram of a geometrical study of abdominal aortas [Figure 1] and computational analysis of blood flow velocity in the idealized abdominal aortas [Figure 2] are presented. Sun and Chaichana describe how the blood flow dynamics have an impact on the formation of plaques in the coronary artery. This analysis may prove a useful starting point in the future development of treatments for plaque rupture and related vascular diseases.[2],[3],[4],[5],[6],[7],[8],[9],[10] By using a computational simulation, their research reveals information about blood flow conditions that cannot be seen by the current traditional medical imaging. The use of intensive computation in medical simulation is also crucial in predicting the long-term outcomes of surgical operations as well as for surgical planning for operations.[9],[10] The medical simulation predicts the effect of stent wires crossing the renal artery on blood flow changes after implanting suprarenal stent graft for the treatment of abdominal aortic aneurysm. Specifically, their research showed how the blood flow velocity changed due to stent wires crossing the renal ostium. Some simulations showed that this led to a dramatic drop in velocity which in turn led to blood flowing into the kidney. Currently, medical imaging can only provide anatomical structures of the blood flow of the lumen artery, and therefore could not predict the consequence of low-velocity blood flow as seen in the simulation. | Figure 1: A schematic diagram of computing approaches in medicine for the geometrical study of abdominal aortas
Click here to view |
 | Figure 2: Computational results of blood flow velocity in idealized abdominal aorta (a), and idealized abdominal aorta with aneurysm (b). Arrows identify locations of low velocity
Click here to view |
Medical modeling is now transforming the static morphological study of human anatomy into a useful computational tool for computer-aided medical simulation and diagnosis. Sourbron and Buckley carried out the medical modeling of a single kidney using dynamic contrast-enhanced magnetic resonance imaging to estimate kidney function.[11],[12],[13],[14] They developed a medical model of the single-kidney compartments to evaluate the healthy conditions of single-kidney glomerular filtration rate. Their work developed medical software that integrates new medical imaging techniques into the sphere of computing in medicine. Such simulations serve as a useful replacement for invasive techniques, such a surgery or blood tests, in order to measure the kidney glomerular filtration rates. Taylor et al. developed a noninvasive computation to evaluate the blood flow reserved at the presence of coronary stenosis using computed tomography.[15] Their work calculates the fractional flow reserve (FFR) derived from coronary computed tomography. This approach has the potential to replace an invasive coronary angiography. Alternatively, their purposed FFR factor can also use to verify an invasive technique for diagnosis of cardiovascular disease.
Three-dimensional (3D) printing can also be used to rapidly create human anatomical systems. These models are useful for education and research but can also support surgical planning. McFarlane and Dentith undertook an anatomical study of a femur head to form a femoral topology for medical modeling. Their work resulted in mathematical geometries that aid understanding of hip pain, hip disorders, and abnormal bone structures, as well as determining the angulation of the femoral neck for surgical planning. Liu et al. used 3D printed models to study geometries of kidney cancer (principally renal tumors).[16] Their work generated realistic tactile models that a surgeon could touch in order to feel the locations of renal tumors. Clearly, this tactile information would be entirely lost by merely displaying 3D models on a screen. Their study reported that the accuracy of 3D printed computer models and the actual shapes of real renal tumors were closely related (within ±2% of tolerance). 3D printing can be used to create accurate physical models where accuracy is judged by comparing the printed solid model with a two-dimensional (2D) image; such comparisons are shown for a femur head and renal tumors [Figure 3] and kidney structure [Figure 4]. | Figure 3: Transparent kidney model (a), 3D printed kidney (b), transparent femur head model (c) and 3D printed femur head (d)
Click here to view |
Both robotics and artificial intelligence (AI) are increasingly seen as having a great potential in the support of medical procedure and diagnosis. Automatic prediction of symptoms, patient advice and support guidance, identification of tumors, and automatic image analysis from tomography and magnetic resonance imaging data have all used AI. Just as important is the way the information is presented to the end user. Chaichana and Sun presented a machine-learning system to classify the similarity of the shapes of the aortic root from the imaging database. A classification of aortic root shapes is shown in [Figure 5]. McHugh and Chaichana programed a robot to move over 2D electrocardiogram maps to characterize heart conditions.[17] Their work also demonstrates that the resulting information can be represented in many different forms or mediums; even the path chosen by the robot can be used to depict a healthy or diseased heart. Huang et al. presented a new approach through a vision-based learning demonstration in which the technique enabled robots to learn how to sew a stent graft for a needle pathway by observation alone.[18] | Figure 5: A classification of aortic root shapes (a and b are classified in relation to the other images c-e)
Click here to view |
In summary, the authors have presented examples suggesting that digital medicine has massive potential in patient support, assessment of diseases, in patient education, and as a consequence, patient autonomy. Medical modeling, computation, and simulation are significant tools for medical research and form the basis of digital medicine. New technologies such as virtual reality and augmented reality, robotics, and machine learning will ultimately drive a new phase of health care. The potential for noninvasive tools to identify and treat diseases in the future will be a critical part of this new era of digital medicine.
Acknowledgment
The authors gratefully acknowledge the Higher Education Innovation Fund (HEIF), Liverpool Hope University for funding and financial support. The authors sincerely thank Miss. Katie McFarlane and Mr. Joseph Dentith for their help in producing 3D-printed femur models.
Conflict of intrerest
None declared.
| References | |  |
| 1. | Matthews MD. Head Strong: How Psychology is Revolutionizing War. Oxford, United Kingdom: Oxford University Press; 2014.  |
| 2. | Sun Z, Chaichana T. An investigation of correlation between left coronary bifurcation angle and hemodynamic changes in coronary stenosis by coronary computed tomography angiography-derived computational fluid dynamics. Quant Imaging Med Surg 2017;7:537-48. |
| 3. | Sun Z, Chaichana T. Computational fluid dynamic analysis of calcified coronary plaques: Correlation between hemodynamic changes and cardiac image analysis based on left coronary bifurcation angle and lumen assessments. Interv Cardiol 2016;8:713-9. |
| 4. | Chaichana T, Sun Z, Jewkes J. Computation of hemodynamics in the left coronary artery with variable angulations. J Biomech 2011;44:1869-78. |
| 5. | Sun Z, Chaichana T. A systematic review of computational fluid dynamics in type B aortic dissection. Int J Cardiol 2016;210:28-31. |
| 6. | Chaichana T, Sun Z, Jewkes J. Impact of plaques in the left coronary artery on wall shear stress and pressure gradient in coronary side branches. Comput Methods Biomech Biomed Engin 2014;17:108-18. |
| 7. | Chaichana T, Sun Z, Jewkes J. Hemodynamic impacts of various types of stenosis in the left coronary artery bifurcation: A patient-specific analysis. Phys Med 2013;29:447-52. |
| 8. | Chaichana T, Sun Z, Jewkes J. Investigation of the haemodynamic environment of bifurcation plaques within the left coronary artery in realistic patient models based on CT images. Australas Phys Eng Sci Med 2012;35:231-6. |
| 9. | Sun Z, Chaichana T. Fenestrated stent graft repair of abdominal aortic aneurysm: Hemodynamic analysis of the effect of fenestrated stents on the renal arteries. Korean J Radiol 2010;11:95-106. |
| 10. | Sun Z, Chaichana T. Investigation of the hemodynamic effect of stent wires on renal arteries in patients with abdominal aortic aneurysms treated with suprarenal stent-grafts. Cardiovasc Intervent Radiol 2009;32:647-57. |
| 11. | Sourbron SP, Buckley DL. On the scope and interpretation of the Tofts models for DCE-MRI. Magn Reson Med 2011;66:735-45. |
| 12. | Sourbron SP, Buckley DL. Tracer kinetic modelling in MRI: Estimating perfusion and capillary permeability. Phys Med Biol 2012;57:R1-33. |
| 13. | Sourbron S, Ingrisch M, Siefert A, Reiser M, Herrmann K. Quantification of cerebral blood flow, cerebral blood volume, and blood-brain-barrier leakage with DCE-MRI. Magn Reson Med 2009;62:205-17. |
| 14. | Sourbron SP, Michaely HJ, Reiser MF, Schoenberg SO. MRI-measurement of perfusion and glomerular filtration in the human kidney with a separable compartment model. Invest Radiol 2008;43:40-8. |
| 15. | Taylor CA, Fonte TA, Min JK. Computational fluid dynamics applied to cardiac computed tomography for noninvasive quantification of fractional flow reserve: Scientific basis. J Am Coll Cardiol 2013;61:2233-41. |
| 16. | Liu D, Sun Z, Chaichana T, Ducke W, Fan Z. Patient-specific 3D printed models of renal tumours using homemade 3D printer in comparison with commercial 3D printer. J Med Imaging Health Inf 2018;8:303-8. |
| 17. | McHugh DD, Chaichana T. Robotic Dance to Characterise Heart Conditions. Undergraduate dissertation. United Kingdom: Liverpool Hope University, Liverpool; 2016. |
| 18. | Huang B, Ye M, Lee S, Yang GZ. A Vision-guided Multi-robot Cooperation Framework for Learning-by-demonstration and Task Reproduction. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS); 2017. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
|
Search Pubmed for