Global Health Journal, June 2020, Volume 4, Issue 2

Contents lists available at ScienceDirect

Global Health Journal

journal homepage: http://www.keaipublishing.com/en/journals/global-health-journal/

REVIEWS

A review of medical artificial intelligence

Rong Liu a,b,⁎, Yan Rong b, Zhehao Peng c
a
Institute of Medical Innovation and Transformation, Puren Hospital of Wuhan, 1 Benxi Road, Wuhan 430081, China
b
School of Medicine, Wuhan University of Science and Technology, No. 1, Huangjia Lake University Town, Wuhan 430065, China
c
Department of Computer Science, College of Engineering, North Carolina State University, Campus Box 8206, 890 Oval Drive, Engineering Building II, Raleigh, NC 27695, USA

A R T I C L E I N F O A B S T R A C T

Article history: Since the concept of “artificial intelligence” was introduced in 1956, it has led to numerous technological innovations
Received 19 September 2019 in human medicine and completely changed the traditional model of medicine. In this study, we mainly explain the
Received in revised form 9 February 2020 application of artificial intelligence in various fields of medicine from four aspects: machine learning, intelligent
Accepted 23 February 2020 robot, image recognition technology, and expert system. In addition, we discuss the existing problems and future
Available online 20 May 2020
trends in these areas. In recent years, through the development of globalization, various research institutions around
the world have conducted a number of researches on this subject. Therefore, medical artificial intelligence has attained
Keywords: significant breakthroughs and will demonstrate wide development prospection in the future.
Medical artificial intelligence
Machine learning
Intelligent robot
Image recognition
Expert system

1. Overview
Artificial intelligence (AI) was coined by John McCarthy at the Dart-
mouth conference in 1956,1 and the field of artificial intelligence originated
from thereon. Kaplan and Haenlein defined AI as “the ability to process ex-
ternal data systematically and learn from it to achieve specific goals and
tasks”.2 AI involves using machines to simulate human thinking processes
and intelligent behaviors, such as thinking, learning, and reasoning, and
aims to solve complex problems that can only be solved by experts.3 As a
branch of computer science, the field of AI mainly studies the following
contents: machine learning, intelligent robot, natural language understand-
ing, neural network, language recognition, image recognition, and expert
system.4
The concept of artificial intelligence in medicine (AIM) originated in the
early 1970s.5 It aimed to increase the efficiency of medical diagnosis and
treatment with the aid of AI systems. After the 1980s, the development of
AIM could be roughly divided into four stages: (1) infancy (1980s): the “de-
cision tree” algorithm was proposed, and artificial neural networks contin-
ued to develop; (2) adolescence (1990s): “expert systems” continued to
mature due to the emergence of support vector machines; (3) coming-of-
age (2000s): the concept of “deep learning” was proposed, and machine
learning became a prominent theme of AIM; and (4) currently, we are in
the “maturation period” (2010s): as the technologies are relatively
advanced. However, the ability to communicate with people still needs to
be improved. Therefore, we are still in the stage of “weak” AI.6
2. Application of AIM
2.1. Machine learning
In 1959, Arthur Samuel coined the term “machine learning” to denote a
category of algorithms and construction of classifiers.7 The algorithm auto-
matically learns through the input data and builds a model based on the
input data to accurately predict new data.8
Subsequently, the algorithms of machine learning experienced many
major breakthroughs: the backpropagation algorithm was proposed in the
early 1960s.9 In 1982, Paul applied the automatic differentiation method
in neural networks.10 In 1986, Ross Quinlan proposed a well-known ma-
chine learning algorithm, called the “decision tree”, which involves classi-
fying data according to the set rules. Tin Kam Ho created an important
algorithm, called the “random forest” (double spatial feature extraction al-
gorithm), based on random subspace method, which was built using the de-
cision tree.11 In 1995, Vladimir invented the support vector machine (SVM)
model. In 2006, Geoffrey Hinton, a leader in the field of deep learning, pro-
posed the deep learning algorithm. Deep learning is actually based on ma-
chine learning, and convolutional neural network (CNN) is one of its
representative algorithms.

⁎ Corresponding author: dr_liurong@whu.edu.cn.

http://dx.doi.org/10.1016/j.glohj.2020.04.002
2414-6447/© 2020 People's Medical Publishing services Co. Ltd. Publishing services by Elsevier B.V. on behalf of KeAi. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
43 Global Health Journal, June 2020, Volume 4, Issue 2
In 2008, according to the standards of Association for the Advancement
of Medical Instrumentation and the British and Irish Hypertension Society,
a novel measurement model of blood pressure based on CNN—
convolutional recurrent neural network-blood pressure (CRNN-BP) was
constructed to solve the problem related to the extraction of pulse wave-
form feature points and low robustness in traditional medicine, and to in-
crease the precision of the model.12
With continuous development of assisted diagnosis technology, a large
data repository is generated during screening, diagnosis, and treatment of
diseases.13 Organizing and analyzing these data in a short duration can be
a challenge for the doctors. Therefore, machine learning is increasingly
used in medicine to help doctors predict diseases and treatment outcomes.
Random forest is one of the most efficient algorithms in machine learn-
ing. In recent times, random forest plays an important role in medicine, par-
ticularly in predicting diseases. Patients with a history of idiopathic
hemorrhagic ulcers may demonstrate a higher incidence of ulcer recur-
rence. The safety of the patient is endangered, if a serious complication,
such as ulcer rupture, occurs. In 2018, machine learning was used to
build a model with high accuracy to predict idiopathic peptic ulcer
rebleeding, which was called IPU-ML.14
In another instance, severe hand-foot-mouth disease caused by entero-
virus can lead to serious complications, such as pulmonary edema and myo-
carditis, in very few children.15 In 2019, to predict severe hand-foot-mouth
disease, CatBoost model was build, which demonstrated higher specificity
and sensitivity than other models, such as decision tree and SVM.16
In addition, machine learning can predict the efficacy of radiotherapy.
For example, patients usually undergo radiation therapy, when they are suf-
fering from lung cancer, especially small cell lung cancer. However, long-
term radiotherapy can lead to serious complications, such as radiation
pneumonitis, which can result in respiratory failure and death.17 Using ar-
tificial neural networks, researchers deduced a method for predicting radi-
ation pneumonitis. They also established a network with an extensive
training in memory and data, which could exhibit higher accuracy in
predicting the complications.18
The “black box” problem of machine learning needs to be resolved.
“Black box” is a neural network comprising CNNs for feature extraction
and recurrent neural networks with long short-term memory. Generally, a
neural network comprises a series of neural layers, including input, process-
ing, and output. The intermediate processing in the neural network is called
the “black box”. It is the system that hides its internal structure from the
user.19 By resolving the problem of black box, the accuracy and computing
power of machine learning can be improved and its application range can
be expanded, which in turn can result in more contributions to the medical
field.
2.2. Intelligent robots
In 1979, the Robot Institute of America defined robot as “a reprogram-
mable, multi-functional manipulator designed to move materials, parts,
tools, or other specialized devices through various programmed motions
for the performance of a variety of tasks”.20 Intelligent robots were used
for surgery in 1980s.21 For example, PUMA 560 was employed in neurosur-
gical biopsy in 1985 and prostate surgery in 1988. ROBODOC, which was
developed in 1992, was the first intelligent robot authorized by the U.S.
Food and Drug Administration (FDA). It was mainly used for hip replace-
ment procedures in orthopedic surgery.
Currently, there are three kinds of robotic surgical systems approved by
the FDA, including the ZUES, Da Vinci, and automated endoscopic system
for optimal positioning robots.22 Owing to their minimally invasive, pre-
cise, and intelligent features,23 intelligent robots are widely used in ortho-
pedics, urology, orthopedics, stomatology, and other fields.
According to the type of orthopedic surgery, robots can be classified
into three categories, including joint surgery robots, spinal orthopedic ro-
bots, and traumatic orthopedic robots.24 Femoral neck fractures can possi-
bly occur in elderly patients with symptoms of hip deformity, pain, and
dysfunction. These fractures can lead to complications, such as fracture
43
nonunion and avascular necrosis of the femoral head, and the best treat-
ment for these fractures is surgery. In 2018, a research explored methods
to reduce the amount of haemorrhagia during the femoral neck fracture sur-
geries. They compared the two surgical methods of orthopedic surgery
robot and manual nailing and concluded that with the assistance of surgical
robots, the surgeon could locate the operative site accurately and reduce
punctures, thus reducing the amount of haemorrhagia during the surgery.25
Intelligent robots are also widely used in gynecological surgery. For ex-
ample, during the early stages of ovarian cancer, patients might suffer from
abdominal mass, ovarian tumor pedicle torsion, and tumor rupture. There-
fore, performing surgical treatments during the early stages of this cancer is
important. A meta-analysis demonstrated that the Da Vinci System had
many advantages during surgery, as it enabled the removal of numerous
lymph nodes and a low blood transfusion rate in the patients. Therefore,
this kind of surgical approach was even safer than the laparoscopic
surgery.26
The robots that assist the surgeons in clinical practice are mostly dis-
crete robots with limited mobility. However, in recent years, continuous ro-
bots (a novel bionic robot with “invertebrate” flexible structure) have been
emerging, and they will gradually replace the discrete ones with their flex-
ible bending characteristics and good adaptability to the environment. Con-
tinuous robots are expected to become the main force of surgery in the
future.27 Although intelligent robots are widely used in the field of orthope-
dics, they still demonstrate drawbacks, such as high costs, large size, and
limited application range.21 With constant advancement of medical and
AI technologies, intelligent robots will be made to gradually adapt to the de-
velopment direction of surgeries in the future.23
2.3. Image recognition technology
Image recognition is the technology of processing and analyzing images
through computers. It is an important technology in the field of AI, which is
based on deep learning. The development of image recognition technology
includes three stages—text recognition, digital image recognition, and ob-
ject recognition.28 The recognition process can be divided into five steps,
including processing the input, image preprocessing, image extraction, clas-
sifier construction, and producing the output.29 This technology can pro-
cess image data rapidly and efficiently. For example, a study found out
that the map projection technology could identify the bones that were
most prone to fractures between femoral and trochanter more efficiently
and accurately than traditional heat map technology.30 They established
that the distribution area of the fracture line was relevant to the age and
gender of the patients.30 Image recognition technology has been highly sig-
nificant in the diagnosis and treatment of intertrochanteric fractures. In ad-
dition, it is also widely used in disease prediction, diagnosis, and lesion
identification.31
Currently, image recognition technology is applied in many clinical dis-
ciplines. Cervical cancer is one of the four major causes of death in
women,32 and it is caused mostly due to infection with human papillomavi-
rus. Patients do not experience any obvious symptoms in the early stages.
However, symptoms, such as anemia and cachexia, may be experienced
later. Although many treatments are available for patients with cervical
cancer, such as surgery, radiotherapy, and chemotherapy, the prognosis
of patients depends highly on whether the cancer is diagnosed at an early
stage. Based on deep learning, intelligent image recognition of the cervix
could assist doctors in the early diagnosis of cervical cancer with an accu-
racy rate of approximately 90%.33
In 2017, a study explored whether the accuracy of diagnosing fungal
keratitis could be enhanced using image recognition technology.34 The re-
searchers adopted a method, called slit lamp microscopy, analyzed the ex-
perimental data, and finally concluded that the detection method based
on image recognition demonstrated higher specificity and sensitivity in
the diagnosis of fungal keratitis than cornea smear. In addition, the image
recognition method could assist doctors with inadequate knowledge in di-
agnosing the disease more accurately.
44 Global Health Journal, June 2020, Volume 4, Issue 2
Deep learning plays an important role in the application of image recog-
nition technology to identify lesions.31 In 2017, a study was using CNNs to
identify malignant breast lesions.35 This method performed better than the
most advanced system in computer-aided detection, as the recognition ac-
curacy of CNN was higher.
Although image recognition technology can aid doctors in diagnosing
diseases in clinical practice, it cannot completely replace the role of doctors.
Due to the different equipment used in varied hospitals, the resolution of
the acquired images can also be different, which can affect the final diagno-
sis to some extent.31 Several difficulties are encountered during the applica-
tion of image recognition technology. For example in multi-layer neural
convolution, the training model requires a large amount of data, and the
computational efficiency needs to be further improved. Besides, high-
performance supercomputers have not yet been popularized. Therefore,
further research in the future is necessary to resolve the problems related
to hardware devices, optimization algorithms, and technology integration.
2.4. Expert system
The expert system is a computer system that simulates the decision-
making ability of human experts.36 As one of the earliest successful AI soft-
ware, it uses the existing knowledge system to reason and solve a series of
complex problems.37
The development phase of the expert system can be roughly divided
into three periods: the initiation period (1965–1971), maturation period
(1972–1977), and the development period (1978–).38 In the early 1960s,
the first expert system—Dendral system was designed. In 1972, the Univer-
sity of Leeds developed the AAPHelp system to assist in the diagnosis of
acute abdomen.39 In 1974, the University of Pittsburgh excogitated the
INTERNIST-I system, which was mainly used for the diagnosis of complex
diseases in internal medicine. In 1976, Stanford University developed the
intelligent diagnosis system MYCIN, which could aid in diagnosing infec-
tious diseases. However, it was not used in clinical practice for various rea-
sons, such as ethics.40
A study stated that the use of personal digital assistants (PDA) to pro-
vide expert knowledge to untrained helpers can significantly improve the
quality of first aid. The evaluation of results concluded that the application
of expert systems could improve the quality of bystanders' first aid, mean-
while strengthen the weakest link of the survival chain.41
In another study, the expert system was used to diagnose different types
of headaches, such as tension headaches, migraine headaches, and drug-
dependent headaches.42 The computerized headache assessment tool
(CHAT) accurately diagnosed 94.4% of migraine headaches and 93% of
daily syndromes. The average accurate rate of diagnosis was 98%. There-
fore, the introduction of the CHAT expert system can aid doctors in identi-
fying the type of headache diagnosed, which is of significant value in health
care.42 In an analysis of the MIT-BIH arrhythmia database, a fuzzy expert
system was adopted to distinguish arrhythmias and ischemic heart beats.
The results demonstrated an average sensitivity of 96% and an average
specificity of 99%.43
The expert system exhibits a strong ability of clinical decision-making
and shows advantages in terms of identification and diagnosis of diseases.
However, it is also necessary to enhance the accuracy of the system, com-
bine the system with the patient's medical history, and simultaneously inte-
grate the doctor's clinical experience. In addition, the knowledge and
findings in medicine must be constantly updated to provide doctors with
frontier diagnosis and treatment plans.44
3. Conclusion
AI is expected to encounter bigger challenges in future development: in
data mining and machine learning, researchers are required to construct a
model of black box predictor to resolve the “black box” problem and invent
the 5th generation wireless technology (5G) and Internet of Things (IoT) in-
tegrated continuous robots; in image recognition technology, a more effi-
cient training model needs to be created, and the expert system should
44
continuously expand the knowledge base to provide more information to
medical personnel.
In the past 10 years, significant breakthroughs have been experienced in
the field of AI. Research institutions in many countries around the world
have demonstrated extensive cooperation in realizing these breakthroughs.
The amount of literature related to AI has developed rapidly at this stage, as
researchers in China, Europe, and the United States have realized signifi-
cant achievements in this field. Particularly, China is gradually becoming
a leader in the field of AI.45
With the high-speed transmission of 5G network, real-time remote tech-
nical guidance for remote collaborative surgery can guarantee the stability,
reliability, and safety of the surgery. It enables experts to understand the
progress of the procedure and the patient's condition in real time, thereby
effectively reducing the risk of the surgery.
AI has completely changed the traditional model of medicine, signifi-
cantly improved the level of medical services, and guaranteed human
health in various aspects. A broader development prospect for medical AI
is highly expected in the future.
Funding
This paper was supported by Hubei Health and Family Planning Com-
mission joint Fund Innovation Team Project (Grant No. WJ2018H0042).
Competing interests
The authors declare that they have no known competing financial inter-
ests or personal relationships that could have appeared to influence the
work reported in this paper.
References
1. Puppe F. Introduction to knowledge systems. Artif Intell Med. 1997;9(2):201–203.
2. Kaplan A, Haenlein M. Siri, Siri, in my hand: who's the fairest in the land? On the inter-
pretations, illustrations, and implications of artificial intelligence. Bus Horiz. 2019;62(1):
15–25.
3. Shen TL, Fu XL. Application and prospect of artificial intelligence in cancer diagnosis and
treatment. Chin J Oncol. 2018;40(12):881–884. [in Chinese].
4. Li SR. Talking about artificial intelligence technology and its application development.
Electron Test. 2019;1:137–138. [in Chinese].
5. Patel VL, Shortliffe EH, Stefanelli M, et al. The coming of age of artificial intelligence in
medicine. Artif Intell Med. 2009;46(1):5-17.
6. Peleg M, Combi C. Artificial intelligence in medicine AIME 2011. Artif Intell Med.
2013;57(2):87–89.
7. Samuel AL. Some studies in machine learning using the game of checkers. IBM J Res Dev.
1959;3(3):211–229.
8. Kumar G, Kalra R. A survey on machine learning techniques in health care industry. Int J
Recent Res Aspect. 2016;3(2):128–132.
9. Schmidhuber J. Deep learning in neural networks: an overview. Neural Netw. 2015;61:
85-117.
10. Werbos PJ. Applications of advances in nonlinear sensitivity analysis. In: Drenick R,
Kozin F, eds. System modeling and optimization. Lecture note in control and information sci-
ences. Berlin, Heidelberg: Springer; 1982.
11. Ho TK. The random subspace method for constructing decision forests. IEEE Trans Pattern
Anal Mach Intell. 1998;20(8):832–844.
12. Zhang JS, Gu LY, Jiang SY. Blood pressure measurement model based on convolutional
recurrent neural network. Beijing Biomed Eng. 2018;37(5):494–501. [in Chinese].
13. Abernethy AP, Etheredge LM, Ganz PA, et al. Rapid-learning system for cancer care. J
Clin Oncol. 2010;28(27):4268–4274.
14. Wong GLH, Ma AJH, Deng HQ, et al. Machine learning model to predict recurrent ulcer
bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding. Aliment
Pharmacol Ther. 2019;49(7):912–918.
15. Liu Z, Wang SK, Yang RS, Ou X. A case-control study of risk factors for severe hand-
footmouth disease in Yuxi, China, 2010–2012. Virol Sin. 2014;29(2):123–125.
16. Wang B, Feng HF, Wang F, et al. Application of CatBoost model based on machine learn-
ing in predicting severe hand-foot-mouth disease. Chin J Infect Control. 2019;18(1):12–
16. [in Chinese].
17. Kong FM, Hayman JA, Griffith KA, et al. Final toxicity results of a radiation-dose escala-
tion study in patients with non–small-cell lung cancer (NSCLC): predictors for radiation
pneumonitis and fibrosis. Int J Radiat Oncol Biol Phys. 2006;65(4):1075–1086.
18. Su M, Miften M, Whiddon C, Sun XJ, Light K, Marks L. An artificial neural network for
predicting the incidence of radiation pneumonitis. Med Phys. 2005;32(2):318–325.
19. Guidotti R, Monreale A, Ruggieri S, Turini F, Pedreschi D, Giannotti F. A survey of
methods for explaining black box models. ACM Comput Surv. 2018;51(5):1-42.
20. Beasley RA. Medical robots: current systems and research directions. J Robot. 2012;2012:
1-14.
45 Global Health Journal, June 2020, Volume 4, Issue 2
21. Guo S, Guo Z. Research progress in orthopedic surgery robot. Med J Chin People Armed
Police Forces. 2018;29(10):987–990. [in Chinese].
22. Wu Q, Zhao YM. Current research and application situation of robot in stomatology. Int J
Stomatol. 2018;45(5):615–620. [in Chinese].
23. Cai SH, Song YW, Cao XY, Fan HY. Current status and research progress of orthopedic ro-
bots. Shandong Med J. 2018;58(44):90–93. [in Chinese].
24. Zhang JY, Ye ZW. The development of orthopedics in the era of intelligent medicine. J
Clin Surg. 2019;27(1):31–32. [in Chinese].
25. Lei CX, Li Q, Li SZ, et al. Comparison of blood loss in patients with femoral neck fracture
undergoing screw fixation with and without robot navigation. Med J Chin People Armed
Police Forces. 2018;29(7):677–679. [in Chinese].
26. Li W, Lang KN, Li J, Wang WL. Meta analysis of safety and short-term therapeutic effect
of Da Vinci robot in early ovarian cancer surgery. J Zhengzhou University (Med Sci).
2018;53(6):811–816. [in Chinese].
27. Sun LN, Hu HY, Li MT. A review on continuum robot. Robot. 2010;32(5):688–694. [in
Chinese].
28. Zhang JY. Current situation and perspective of image recognition technology. Computer
Knowledge Technol. 2010;6(21):6045–6046. [in Chinese].
29. Masood A, Sheng B, Li P, et al. Computer-assisted decision support system in pulmonary
cancer detection and stage classification on CT images. J Biomed Inform. 2018;79:117–
128.
30. Fu YC, Liu R, Liu Y, Lu JW. Intertrochanteric fracture visualization and analysis using a
map projection technique. Med Biol Eng Comput. 2019;57(3):633–642.
31. Liu F, Zhang JR, Yang H. Research progress of medical image recognition based on deep
learning. Chin J Biomed Eng. 2018;37(1):86–94. [in Chinese].
32. International Agency for Research on Cancer. World cancer report 2014. Geneva: WHO;
2015.
45
33. Liu J, Sun RC, Qiao SL. Application and research of deep learning in medical image rec-
ognition. J Qingdao Univers (Nat Sci Edit). 2018;31(1):69-74,80. [in Chinese].
34. Wu XL, Tao Y, Qiu QC, Wu XY. Application of image recognition-based automatic hyphae
detection in fungal keratitis. Australas Phys Eng Sci Med. 2018;41(1):95-103.
35. Kooi T, Litjens G, van Ginneken B, et al. Large scale deep learning for computer aided de-
tection of mammographic lesions. Med Image Anal. 2017;35:303–312.
36. Myers W. Introduction to expert systems. IEEE Expert. 1986;1(1):100–109.
37. Russell SJ, Norvig P. Artificial intelligence: a modern approach. 3rd ed. London: Pearson;
2009.
38. Liu JK, Deng SQ. History and trends of expert system development. Comput Era. 1995;5:
20–21. [in Chinese].
39. Sim I, Gorman P, Greenes RA, et al. Clinical decision support systems for the practice of
evidence-based medicine. J Am Med Inform Assoc. 2001;8(6):527–534.
40. Shortliffe EH, Buchanan BG. A model of inexact reasoning in medicine. Math Biosci.
1975;23(3–4):351–379.
41. Ertl L, Christ F. Significant improvement of the quality of bystander first aid using an ex-
pert system with a mobile multimedia device. Resuscitation. 2007;74(2):286–295.
42. Maizels M, Wolfe WJ. An expert system for headache diagnosis: the Computerized Head-
ache Assessment Tool (CHAT). Headache. 2008;48(1):72–78.
43. Exarchos TP, Tsipouras MG, Exarchos CP, Papaloukas C, Fotiadis DI, Michalis LK. A
methodology for the automated creation of fuzzy expert systems for ischaemic and ar-
rhythmic beat classification based on a set of rules obtained by a decision tree. Artif Intell
Med. 2007;40(3):187–200.
44. Sheikhtaheri A, Sadoughi F, Dehaghi ZH. Developing and using expert systems and neural
networks in medicine: a review on benefits and challenges. J Med Syst. 2014;38(9):110.
45. Tran BX, Vu GT, Ha GH, et al. Global evolution of research in artificial intelligence in
health and medicine: a bibliometric study. J Clin Med. 2019;8(3):360.