Scope of Artificial Intelligence in Medicine

Atul Dwivedi¹, Shweta Shukla Dwivedi^2*, Muhammad Raheel Tariq³, Xiaoming Qiu⁴, Suzhen Hong¹ and Yu Xin¹

^*Correspondence: Shweta Shukla Dwivedi, Department of Dental Surgery, India, Email:

Author info »

Introduction

Artificial Intelligence (AI), first proposed by Professor John McCarthy in 1955, direct to reproduce human intelligence using computers and other electronic devices [1]. The concept of Computer Aided Diagnosis (CAD) was introduced in 1980. Its primary intention was to provide second opinion to radiologists [2]. CAD is strongly related to Artificial Intelligence (AI). CAD and automated computer diagnosis (ACD) are two concepts with different meanings [3].

AI concept has become tremendously popular in recent years, as this concept has empowered search engines, automatic vehicles, voice recognition software’s etc. have mesmerized our daily lives. Moreover, AI also gets importance and shows great promise in improvement of research quality, practice efficacy, personalized patient management, drug effects etc. [4,5].

Artificial Intelligence has revolutionized the medical world and can be commonly regarded as the component of computer sciences, that's is able to handle huge data easily with little theory [6]. AI-powered medical technologies enable a 4P model of medicine, that is predictive, preventive, personalized, participatory [7]. Smartphones are filling the gap and distribute the electronic personal health record [8], helping patients for monitoring vital functions with biosensors and to reach optimal therapeutic compliance [9,10]. AI techniques can help the clinicians to sort out clinically relevant information from massive amount of data, so that clinicians can take decisions easily and appropriately. Additionally, AI can assist the researchers to analyses huge clinical data within a short time, in order to make the healthcare facility successful [11-13].

In other words, Artificial Intelligence also assist physicians by giving them updated information from journals, websites, e-books about clinical practice. Simultaneously, Artificial Intelligence System can minimize the diagnostic and therapeutic errors, which are inevitable to avoid. With real time reference, as technology moving ahead, Artificial Intelligence filters key information from gigantic population data to make prior health risk alert and therapeutic outcome predictions [13-17].

Current Applications of Artificial Intelligence(AI) in Medicine

AI is becoming the main constituent of health care, including drug discovery, remote patient monitoring, medical diagnostics, medical imaging, virtual assistance, hospital management and risk assessment and management. Many branches of medicine dealing with huge data such as analysis of DNA and RNA sequencing, imaging data including radiology, pathology, dermatology and ophthalmology, have already benefitted from Artificial Intelligence implementations [18-20]. Pathologists have utilized Artificial Intelligence to decrease their error rate in the recognition of cancer- positive lymph nodes from 3.4% to 0.5% [21]. Additionally, Artificial Intelligence improved the outcomes in the field of surgery and radio-diagnosis for improved identification of high risk patients, also reduced the rate of lumpectomy by 30% in patients whose breast needle biopsies are considered high risk lesions but finally found to be benign after surgical excision [22].

Recently, Bhandari et al. reported that the progress in cloud computing, big data analytics and AI have led to evolution of more intelligent robots and subsequently with the application of deep learning, several surgical companies are collaborating with tech Companies for developing intelligent robots. There is huge development in the field of autonomous robotics. However, the role of robotics in operation theatre remains hazy. It is time, when surgeons actively take part in development of next generation smart and intelligent robots [23].

The primary aim behind the emergence of Artificial Intelligence in radiodiagnosis is because of greater efficacy and efficiency in clinical outcome & clinical settings. Patrick et al. also reported that Magnetic Resonance Imaging (MRI) is an effective modality in the selecting the patients suitable for arthroscopy from a general population [24]. Google launched its Deep mind Health project, which is used to collect the data of medical records for efficient health services. In 2016, they also launched a cooperative project with Moorefield’s Eye Hospital NHS Foundation Trust to improve eye treatment [25]. Some other companies also there, who are using artificial intelligence to improve health care outcome mentioned in Table 1.

Table 1: Shows List of companies using AI to improve outcome in health services.

Study reported that AI and ML can be used in the diagnosis of interstitial lung disease and few studies also reported about interpretation of images on chest X-ray and diagnosis of bronchial asthma. Despite, ML may be helpful in making clinical decisions, can't replace the expertise completely [26]. Importantly, Nick identified 16 review studies related with AI applications in acute stroke diagnosis and added that stroke onset timing, core and penumbral volume, clinical examination may be combined for exact detection of LVO (Large vessel occlusion) in order to enhance the patient selection for a rapid thrombectomy [27].

Applications of Artificial Intelligence have also shown importance in prediction of development of Alzheimer's disease from positron emission tomography [28], detecting cancers in mammograms [29], analyzing CT scans [30,31], identifying brain tumor lesions Magnetic resonance imaging [32], detecting Arrhythmias [33], identification of cancerous skin lesions [34,35], assessment of embryo quality for maximizing the success chances of in vitro fertilization [36], interpretation of retinal imaging [37], improvement in genomics expression [38] etc.

Commonly, Outcomes of most of the orthopedic surgeries are uncertain, because even after success full surgeries, surgeons can't avoid inevitable complications such as malunion, implant failure, surgical wound infections, Heterotopic ossification, spinal trauma & hematoma in Artificial Cervical Disc Replacement (ACDR), etc. Another study concluded that in case of pilon fracture management, no single method or approach can be regarded as ideal for the treatment of pilon fracture, even surgeon's skills and choice also matter a lot in the outcome of surgery. Therefore, we need to implement stem cell regenerative medicine (SCRM) in place of surgeries. We can regenerate several damaged tissues, tendons, cartilages, muscles, etc. [39-42]. In this scenario, AI techniques could be used for optimization of stem cell therapy and gene therapy in Pediatric patients by predicting clinical outcomes, simplifying cost and treatment [43]. In near future, hopefully researchers can implement the same in all branches of medicine.

Limitation of artificial intelligence

Artificial Intelligence also has its own limitations, as forecasting and predictions about outcomes of therapeutics may be wrong in novel cases of viral infections ( e.g. Novel corona virus infection or COVID -19, recently we have limited information about behavior of this virus and definitive treatment modalities ), novel side effects or resistant cases, newly reported side effects, rare disease, novel modalities of treatment, where there is no prior information for therapeutics, diagnosis, prognosis and outcomes. Obviously, in the lack of information, AI may not replace human knowledge and human brain reasoning power [44]. Telemedicine filling the gap between Multispecialty hospital and Primary health care units, where Surgeons and super specialist can give relevant suggestions and diagnosis online, but they can't provide surgical treatment online.

Conclusion

Artificial Intelligence is expanding its roots in several branches of medicine, such as radiology, pathology, oncology, surgery, etc. As technology evolves, Artificial Intelligence also able to change the decision of Surgeons and Physicians. Additionally, healthcare professional able to keep themselves safe in the era of pandemic (COVID -19), with well programmed robots, which can scan the patient's vital sign without coming into direct contact with them. Last but not the least, huge clinical evaluation is essential for ensuring the efficiency, accuracy and safety of Artificial Intelligence.

References

1. Deo RC. Machine learning in medicine. Circulation 2015; 132:1920-1930.
2. Huang HK. Medical imaging, PACS, and imaging informatics: Retrospective. Radiological Phy Technol 2014; 7:5-24.
3. Mokli Y, Pfaff J, dos Santos DP, et al. Computer-aided imaging analysis in acute ischemic stroke-background and clinical applications. Neurol Res Practice 2019; 1:2.
4. Panchmatia JR, Visenio MR, Panch T. The role of artificial intelligence in orthopaedics surgery. Br J Hosp Med 2018; 79:676-681.
5. Track C. Artificial intelligence and machine learning: Applications in musculoskeletal physiotherapy. Musculoskelet Sci Pract 2019; 39:164-169.
6. Peng Y, Zhang Y, Wang L. Artificial intelligence in biomedical engineering and informatics: an introduction and review. Artif Intell Med 2010; 48:71-73.
7. Orth M, Averine M, Chatzipanagiotou S, etal. Opinion: Redefining the role of the physician in laboratory medicine in the context of emerging technology, personalised medicine and patient autonomy ('4P medicine'). J Clin Pathol 2019; 72:191-197.
8. Abdulnabi M, Al- Haiqi A, Kiah MLM, et al. A distributed framework for health information exchange using smartphone technologies. J Biomed informat 2017; 69:230-250.
9. Topol EJ. A decade of digital medicine innovation. Sci Trans Med 2019; 11:7610.
10. Morawski K, Ghazinouri R, Krumme A, et al. Association of a smartphone application with medication adherence and blood pressure control: The MedISAFE-BP Randomized clinical trial. JAMA Int Med 2018; 178:802-809.
11. Murdoch TB, Detsky AS. The inevitable application of big data to health care. JAMA 2013; 309:1351-1352.
12. Kolker E, Özdemir V, Kolker E. How Healthcare can refocus on its super customers (Patient, n=1) and customers (Doctors and Nurses) by leveraging lessons from amazon, uber, and Watson. OMICS 2016; 20:329-33.
13. Dilsizian SE, Siegel EL. Artificial intelligence in medicine and cardiac imaging: Harnessing big data and advanced computing to provide personalised medical diagnosis and treatment. Curr Cardiol Resp 2014; 16:441.
14. Neil DB. Using artificial intelligence to improve hospital inpatient care. Intell Syst 2013; 28:92-95.
15. Patel VL, Shortliffe EH, Stefanelli M, et al. The coming of age of artificial intelligence in medicine. Artif Intell Med 2009; 46:5-17.
16. https://www.ibm.com/developerworks/community/blogs/InsideSystemStorage/entry/ibm_watson_how_to_build_your_own_watson_jr_in_your_basement7?lang=en
17. Lee CS, Nagy PG, Weaver SJ, et al. Cognitive and system factors contributing to diagnostic errors in radiology. Am J Roentgenology 2013; 201:611-617.
18. Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fund us photographs. JAMA 2016; 316: 2402-2410.
19. Rusk N. Deep learning. Nature Methods 2015; 13:35.
20. Esteva A, Kuprel B, Novoa RA, et al. Dermatologist: Level classification of skin cancer with deep neural networks. Nature 2017; 542:115-118.
21. https://arxiv.org/abs/1606.05718
22. Bahl M, Barzilay R, Yedida AB, et al. High-risk breast lesions: A machine learning model to predict pathologic upgrade and reduce unnecessary surgical excision. Radiology 2018; 286:810-818.
23. Bhandari M, Zeffiro T, Reddiboina M. Artificial intelligence and robotics surgery: Current perspective and future directions. Curr Opin Urol 2020; 30:48-54.
24. Vincken PW, Braak BP, van Erkell AR, et al. Effectiveness of MR imaging in selection of patients for arthroscopy of the knee. Radiology 2002; 223:739-746.
25. https://www.moorfields.nhs.uk/news/moorfields-announces-research-partnership
26. Mekov E, Miravitlles M, Petkov R. Artificial intelligence and machine learning in respiratory medicine: Expert Rev Respir Med 2020; 17:1-6.
27. Murray N. Artificial intelligence in acute stroke diagnostics: Application in large vessel occlusions. Neurology 2019; 92:6.
28. Ding Y, Sohn JH, Kawczynski MG, et al. A deep learning model to predict a diagnosis of alzheimer disease by using 18F-FDG PET of the Brain. Radiology 2019; 290:456-464.
29. https://arxiv.org/abs/1703.07047
30. Hua KL, Hsu CH, Hidayati SC, et al. Computer aided classification of lung nodules on computed tomography images via deep learning technique. Onco Targets Ther 2015; 8:2015-2022.
31. https://ieeexplore.ieee.org/document/8363515
32. Kamnitsas K, Ferrante E, Parisot S, et al. Deep medic for brain tumor segmentation In: international workshop on brainlesion: Glioma, Mutiple sclerosis, stroke, and traumatic brain injuries 2016; 138-149.
33. Hannun AY, Rajpurkar P, Haghpanahi M, et al. Cardiologist: Level arrhythmia detection and classification in ambulatory electrocardiograph using a deep neural network. Nat Med 2019; 25:65-69.
34. Han SS, Kim MS, Lim W, et al. Classification of the clinical images for benign and malignant cutaneous tumors using a deep learning algorithm. J Invest Dermatol 2018; 138:1529-1538.
35. Brinker TJ, Hekler A, Enk AH, et al. Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task. Eur J Cancer 2019; 113:47-54.
36. Khosravi P, Kazemi E, Zhan Q, et al. Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilisation. Digit Med 2019; 2:21.
37. De Fauw J, Ledsam JR, Romera-Paredes B, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med 2018; 24:1342-1350.
38. Xu J, Yang P, Xue S, et al. Translating cancer genomics in to precision medicine with artificial intelligence: Applications, challenges and future perspective. Hum Genet 2019; 138:109-124.
39. Dwivedi A, Dwivedi SS, Su Zhenhong, et al. Open reduction and Internal Fixation of posterior pilon variant fractures with buttress plate through posterolateral approach. IJCMR 2019; 5:1-5.
40. Dwivedi A, Dwivedi SS, Tariq MR, et al. Stem cell regenerative medicine (SCRM)-A new hope in orthopedics-Review article. J Stem Cell Biol Transplant 2019; 3:1-4.
41. Dwivedi A, Jian WX, Dwivedi SS, et al. Pilon fracture: An unsolved riddle: An updates review. IJCMR 2017; 4:718-725.
42. Dwivedi A, Jian WX, Dwivedi SS. Artificial cervical disc replacement: A double edged sword-A clinical review. IJCMR 2017; 4:1163-1168.
43. Sniecinski I, Seghatchian J. Artificial intelligence: A joint narrative on potential use in Pediatric stem and immune cell therapies and regenerative medicine. Transfus Apher Sci 2018; 57:422-424.
44. Mesko B. The role artificial intelligence in precision medicine. Expert Review Precision Med Drug Develop 2017; 2:239-241.