JIS University

Research Paper
THE FUTURE OF MEDICINE WITH ARTIFICIAL INTELLIGENCE
SUHANA SAPUI, SUPRIYA GHOSH
1
Department of Computer Science Engineering, JIS University, KOLKATA, India
2
Department of Computer Science Engineering, JIS University, KOLKATA, India

Email address: suhanasapui2003@gmail.com (Suhana Sapui), supriyohoney@gmail.com (Supriya Ghosh)

Received: Accepted: Published: DOI:

Abstract: Artificial Intelligence in Medicine (AIM) has experienced significant advancements over the past five decades,
particularly with the emergence of Machine Learning (ML) and Deep Learning (DL). These AI methods have expanded the
application of AIM, offering new possibilities for personalized medicine compared to traditional algorithm-based approaches.
Deep Learning techniques enable the analysis of vast amounts of data obtained from smart gadgets, such as smartphones,
wearables, and mobile monitoring sensors, across various medical domains. With ongoing improvements in computational
processing, AI is enhancing the accuracy and efficiency of diagnosis and treatment in various medical specializations. While
patients embrace the implementation of augmented medicine for its autonomy and personalized treatment options, physicians
still exhibit hesitancy due to the integration of these modern tools without traditional clinical trials. This paper explores the
future of medicine with Artificial Intelligence, discussing its potential impact and challenges.

Keywords: artificial intelligence, machine learning, deep learning, medicine, patient, physician

© 20XX, IJCSE All Rights Reserved 1

1. INTRODUCTION mitigate risks, this paper aims to facilitate the successful
integration of AI in healthcare, ultimately improving patient
Medical technology encompasses a diverse range of tools outcomes and revolutionizing clinical practice.
used by healthcare professionals to improve patient care,
enhance early diagnosis, reduce complications, optimize The rest of the paper is organized as follows, section 2 is
treatment, and minimize hospitalization duration. about the comparison of artificial intelligence with deep
Historically, medical technology primarily referred to classic learning and machine learning, Section 3 is about the
devices such as prosthetics, stents, and implants. However, application of AI in different branches of medical sciences,
the advent of smart phones, wearable devices, and Section 4 is about the impact of AI on the data privacy of the
communication systems has ushered in a new era, patients, Section 5 is about how AI is transforming the
revolutionizing medicine with the integration of artificial current scenario of medicine,Section 6 is a case study of the
intelligence (AI) powered tools in compact sizes. This paper related topic, lastly Section 7 is on the impact of AI on
explores the evolution of medical technology, from traditional physicians.
devices to AI-powered tools, highlighting their role in
improving healthcare outcomes and patient quality of life.
2. ARTIFICIAL INTELLIGENCE VS
1.1 Artificial Intelligence: MACHINE LEARNING VS DEEP LEARNING
Artificial intelligence (AI) is rapidly gaining significance Artificial intelligence (AI) is a subfield of computer science
in medicine and healthcare due to advancements in focused on designing intelligent machines that can mimic
computing power, learning algorithms, and the human cognitive functions and perform tasks in an intelligent
availability of large datasets sourced from medical manner. In the field of medicine, AI has undergone
records and wearable health monitors. The increased significant evolution. Initially, expert systems attempted to
availability of Graphic Processor Units and infinite encode physician decisions into rule-based algorithms.
computing resources in the cloud has further accelerated However, the complexity and variability of diseases made it
the growth of AI in healthcare. Learning algorithms are challenging to create a comprehensive set of rules for all
becoming more precise and accurate, enabling valuable clinical situations. Subsequently, machine learning (ML)
insights into diagnostics, treatment options, and patient emerged as a substitute for expert systems in the 1990s,
outcomes. The rise of AI, particularly in the era of large emphasizing the learning aspect of intelligence.
data, holds great potential for improving patient care,
enhancing diagnostic and treatment efficiency, and ML involves the development of computer programs that can
providing decision support to mitigate errors. This has learn and improve from experience without explicit
led to the development of augmented medicine, which programming. Unlike traditional computer programs, which
employs new medical technologies to improve various require specific instructions for each step, ML algorithms
aspects of clinical practice. While several AI-based learn to detect patterns from data provided during training
algorithms have gained approval from regulatory bodies sessions. Inspired by the structure of the human nervous
like the Food and Drug Administration, augmented system, ML research has focused on creating artificial neural
medicine faces resistance from healthcare professionals, networks (ANNs) consisting of interconnected artificial
particularly physicians. neurons. The introduction of deep neural networks (DNNs)
with multiple layers has led to improved prediction
Augmented medicine, fueled by advancements in capabilities, particularly as the size of the training dataset
artificial intelligence (AI), has demonstrated success in increases.
improving patient outcomes. However, healthcare
professionals, particularly physicians, have exhibited Deep learning, a subfield of ML, utilizes algorithms such as
resistance toward embracing this transformative field. DNNs and convolutional neural networks (CNNs). DNNs can
These reasons include unpreparedness for digital act not only as classifiers but also as feature extractors,
medicine due to inadequate education, increased enabling end-to-end training. CNNs, primarily applied to
administrative burden from healthcare digitalization image segmentation and classification, have benefited from
leading to physician burnout, fear of AI replacing the computational power of Graphics Processing Units
physicians, and the absence of a legal framework (GPUs), significantly accelerating processing speed compared
defining liability in the context of algorithm to traditional CPU-based processing.
recommendations. By acknowledging these concerns and
providing potential solutions, this paper aims to facilitate
the acceptance and adoption of augmented medicine,
fostering collaboration between healthcare professionals 3. AI IN MODERN MEDICINE
and AI technologies for improved patient care.
AI has the potential to bring several benefits to healthcare,
This paper aims to provide a comprehensive overview of including:
recent advancements in AI in medicine, highlighting
successful cases of AI-based medical technologies in clinical Diagnostic Accuracy:
practice. It also delves into the challenges and risks faced by AI algorithms can analyze medical images, test
healthcare professionals while adopting augmented medicine. results, and patient data to assist in diagnosing diseases and
By understanding these challenges and offering strategies to conditions. By leveraging pattern recognition and data

© 20XX, IJCSE All Rights Reserved 2

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

analysis, AI can help improve diagnostic accuracy and reduce effectiveness in identifying atrial fibrillation in
errors. ambulatory patients. Studies, such as REHEARSE-
AF, have shown that remote monitoring with AI-
Efficiency in Provider Workflow: based tools can be more effective than routine care
AI-powered tools can automate repetitive tasks, in detecting cardiac arrhythmias.
streamline administrative processes, and enhance efficiency
in healthcare workflows. This allows healthcare providers to Predictive Analytics for Cardiovascular Risk: AI has
focus more on patient care and spend less time on been employed to analyze electronic patient records
administrative tasks. and predict the risk of cardiovascular diseases,
including acute coronary syndrome and heart failure.
Clinical Operations: AI algorithms can leverage large datasets to identify
AI can optimize clinical operations by analyzing patterns and risk factors, leading to more accurate
large datasets, identifying trends, and providing risk assessments compared to traditional scales.
insights to improve resource allocation, scheduling,
and patient flow. This helps hospitals and healthcare Limitations and Challenges: Wearable and portable
facilities operate more efficiently and effectively. ECG technologies have some limitations, such as
false positives caused by movement artifacts.
Disease and Therapeutic Monitoring: Additionally, the adoption of wearable technology in
Artificial Intelligence can monitor patient data in elderly patients, who are more likely to suffer from
real time, which can help healthcare providers detect atrial fibrillation, may face barriers due to usability
any significant changes or abnormalities. This early concerns or limited familiarity with technology.
detection can lead to the personalized treatment
adjustments, improved disease management, and Sample Size and Research Variability: It's important
better patient outcomes. to note that research results in cardiovascular AI can
vary depending on the sample size used.
Procedure Accuracy: Comprehensive reviews have highlighted the need
AI can assist in surgical procedures by providing for robust study designs and large-scale datasets to
real-time guidance, image analysis, and predictive ensure reliable and consistent predictions.
modeling. Robotic surgery systems, for example,
enhance precision and control during complex AI's integration in cardiology holds great potential
surgeries, leading to improved procedure accuracy for improving patient outcomes, facilitating early
and patient outcomes. detection, and enhancing risk prediction. However,
further research, validation, and addressing technical
Overall Patient Outcomes: limitations are crucial to ensuring the accuracy and
By leveraging AI for personalized medicine, effectiveness of AI algorithms in cardiovascular
treatment planning, and predictive analytics, care.
healthcare providers can tailor treatment approaches
to individual patients, resulting in better patient 3.2. PULMONARY MEDICINE
outcomes and improved quality of care.The
progressive growth and development of the AI AI applications in the interpretation of pulmonary
platform in medicine is categorized below: function tests (PFTs) have shown promise in
pulmonary medicine. PFTs are crucial diagnostic
tests used to assess lung function and aid in the
3.1. CARDIOLOGY diagnosis and management of various respiratory
conditions. Here are some key points regarding the
In the field of cardiology, AI has shown promising development of AI applications in this field:
applications in early detection, remote monitoring, risk
prediction, and disease management. Here are some key Improved Interpretation Accuracy: AI-based
points related to AI's impact in cardiology: software and algorithms have demonstrated the
Early Detection of Atrial Fibrillation: AI algorithms potential to provide more accurate interpretations of
have been utilized in wearable devices, such as the PFT results. By analyzing large datasets and
Apple Watch, to detect atrial fibrillation by leveraging machine learning techniques, AI
analyzing ECG data. These devices enable easy algorithms can learn patterns and associations within
acquisition of ECGs and provide a means for the data, leading to enhanced accuracy in
individuals to monitor their heart rhythm and share interpreting test results.
data with healthcare practitioners.
Decision Support Tool: AI in the context of PFT
Remote ECG Monitoring: Remote ECG monitoring interpretation can serve as a valuable decision
with devices like Kardia has demonstrated support tool for healthcare professionals. It can assist
in the identification and classification of abnormal

© 20XX, IJCSE All Rights Reserved 3

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

lung function patterns, enabling physicians to make the day. This continuous feedback allows for timely
more informed diagnostic and treatment decisions. adjustments in medication, diet, and lifestyle,
leading to improved glycemic control.
Integration of Multiple Parameters: PFTs involve the
measurement of various lung function parameters, Reduction of Stigma: CGM systems help reduce the
such as spirometry, lung volumes, and diffusing stigma associated with hypoglycemic episodes. By
capacity. AI algorithms can integrate and analyze providing early warnings and notifications, patients
these multiple parameters simultaneously, can take proactive measures to prevent severe
considering complex interactions between them. hypoglycemia, thereby promoting a sense of security
This holistic approach can contribute to more and reducing potential embarrassment or social
comprehensive and accurate interpretations. consequences.

Clinical Applications: AI-based PFT interpretation Emotional Impact and Self-Perception: It's important
tools have the potential to aid in the diagnosis and to recognize that managing diabetes can have
monitoring of respiratory conditions, such as chronic emotional implications for patients. While CGM
obstructive pulmonary disease (COPD), asthma, systems offer valuable insights and support, some
interstitial lung diseases, and more. They can individuals may experience feelings of personal
provide insights into disease severity, progression, failure or frustration if they struggle to regulate their
and treatment response, facilitating personalized glucose levels. Patient education, counseling, and
patient care. support systems can play a crucial role in addressing
these emotional aspects and promoting a positive
It's important to note that while AI shows promise in self-perception.
PFT interpretation, it should be considered as a
supportive tool rather than a replacement for Understanding patient experiences and emotions
healthcare professionals. The expertise and clinical related to glucose monitoring is vital for healthcare
judgment of physicians remain crucial in the overall providers to provide holistic care and support. Open
management of patients. Further research and communication, patient education, and addressing
validation are necessary to refine AI algorithms, emotional well-being alongside glucose management
ensure generalizability across diverse populations, can help individuals better navigate the challenges
and address potential challenges or limitations in associated with diabetes management
real-world clinical settings.
3.4. NEPHROLOGY
3.3. ENDOCRINOLOGY
Artificial intelligence (AI) has indeed shown
Continuous glucose monitoring (CGM) systems promise in various applications within clinical nephrology.
have significantly improved the management of Here are some key points regarding the use of AI in
diabetes by providing real-time information on nephrology and the limitations associated with sample sizes:
glucose levels. Here are some key points regarding
the use of CGM and the associated patient Prediction of Decline in Glomerular Filtration Rate
experiences: (GFR): AI algorithms have been utilized to predict
the decline of glomerular filtration rate in patients
Real-Time Glucose Readings: CGM systems, such with polycystic kidney disease. By analyzing various
as the Medtronic Guardian system, allow patients to clinical and imaging data, AI models can assist in
continuously monitor their interstitial glucose levels identifying patients at higher risk of disease
in real time. This provides valuable information on progression. This early prediction can help clinicians
glucose trends, direction, and rate of change, intervene with appropriate management strategies.
enabling patients to make informed decisions about
their diabetes management. Risk Assessment in IgA Nephropathy: AI has also
been applied to establish the risk of progressive IgA
Smartphone Integration: The integration of CGM nephropathy. By analyzing clinical and
systems with smartphones enhances the convenience histopathological data, AI models can identify
and accessibility of glucose monitoring. Patients can factors that contribute to disease progression and
easily view their glucose readings, trends, and alerts help guide treatment decisions. This can potentially
on their smartphones, promoting seamless and on- improve patient outcomes by facilitating
the-go management. personalized care plans.

Optimization of Blood Glucose Control: CGM Limitations of Sample Size: While AI shows
empowers patients to optimize their blood glucose promise in nephrology, one of the current limitations
control by providing a more comprehensive is the requirement for large sample sizes to draw
understanding of their glucose patterns throughout meaningful inferences. The availability of extensive

© 20XX, IJCSE All Rights Reserved 4

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

and diverse datasets is crucial for training accurate

and reliable AI models. However, collecting and Outcome Prediction: AI models, including machine
curating large-scale datasets in nephrology can be learning algorithms, have been utilized to predict
challenging due to the complexity and heterogeneity outcomes in gastrointestinal conditions. For
of renal diseases. instance, these models can predict outcomes in
gastrointestinal bleeding, survival rates in
Need for Collaboration and Data Sharing: To esophageal cancer, disease lammatory bowel
overcome the limitation of sample size, disease, and metastasis in colorectal cancer and
collaboration among research institutions and data esophageal squamous cell carcinoma. By integrating
sharing initiatives are essential. By pooling together multiple patient variables, AI algorithms can provide
data from multiple sources, researchers can create valuable prognostic information for personalized
more robust and generalizable AI models. treatment planning.
Collaborative efforts can help address the scarcity of
data for certain rare renal conditions and improve the Treatment Optimization: AI applications in
overall accuracy and reliability of AI applications in gastroenterology extend to treatment optimization as
nephrology. well. AI algorithms can assist in determining the
most appropriate treatment strategies for individual
Future Directions: As the field of AI continues to patients based on their specific clinical
advance and more data becomes available, the characteristics, optimizing medication selection,
application of AI in nephrology is expected to grow. endoscopic interventions, and surgical approaches.
Efforts to standardize data collection, develop
interoperable electronic health records, and establish It is important to note that while AI shows great
data sharing networks will be crucial in expanding potential in gastroenterology, further validation and
the application of AI and improving the precision of integration into clinical practice are necessary.
predictions and risk assessments in nephrology. Large-scale studies, data standardization, and
collaboration between gastroenterologists and data
While AI holds great potential in nephrology, it is scientists are crucial for refining AI models and
important to validate and refine AI models through ensuring their reliability and accuracy. By
rigorous research and larger-scale studies. leveraging AI technologies, gastroenterologists can
Collaboration among researchers, clinicians, and improve diagnostic accuracy, enhance treatment
data scientists will be key to harnessing the full outcomes, and ultimately provide better care to
potential of AI in improving outcomes for patients patients with gastrointestinal disorders.
with renal diseases.
3.6. NEUROLOGY
3.5. GASTROENTEROLOGY
Intelligent seizure detection devices, such as the
The field of gastroenterology has witnessed the Embrace by Empatica, offer a promising solution for
application of various AI technologies in clinical settings, continuous ambulatory monitoring of epilepsy
leading to advancements in diagnosis, prognosis, and patients. The Embrace wearable, equipped with
treatment. Here are some key AI applications in electrodermal captors, has received FDA approval
gastroenterology: for its ability to detect generalized epilepsy seizures.
It works in conjunction with a mobile application
Image Analysis: Convolutional neural networks that can alert close relatives and trusted physicians,
(CNNs) and other deep learning models have been providing additional information about the patient's
used to analyze images obtained from endoscopy location during a seizure.
and ultrasound in order to detect abnormal
structures, such as colonic polyps. AI algorithms can In terms of patient experience, a report highlighted
assist gastroenterologists in the identification and that epilepsy patients showed a high interest in using
classification of lesions, aiding in early detection and seizure detection devices and did not face any
more accurate diagnosis. significant barriers to adoption. Unlike heart
monitoring wearables, which may have certain
Disease Diagnosis: Artificial neural networks challenges for users, patients with epilepsy were
(ANNs) have been employed to diagnose various more receptive to utilizing wearable technology for
gastrointestinal conditions. For example, ANNs have seizure detection.
shown promise in diagnosing gastroesophageal
reflux disease (GERD) and atrophic gastritis by The availability of these devices can offer peace of
analyzing patient data and identifying characteristic mind to patients and their families, as they provide
patterns associated with these conditions. AI-based continuous monitoring and timely alerts in the event
diagnostic models can support clinicians in making of a seizure. By leveraging AI and wearable
more accurate and efficient diagnoses.

© 20XX, IJCSE All Rights Reserved 5

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

technology, intelligent seizure detection devices more accurate diagnoses. For example, in the case of
have the potential to improve the management and chronic kidney damage, DL algorithms have been
care of epilepsy patients, enhancing their safety and used to predict kidney function and assess the rate of
overall quality of life decline. The results have shown that DL algorithms
outperform traditional pathologist-estimated scoring
3.7. DIAGNOSTIC RADIOLOGY systems in predicting kidney decline, demonstrating
their potential to enhance clinical decision making
AI and ML technologies have shown great potential for nephrologists.
in analyzing large datasets and extracting valuable
insights in the fields of radiology and pathology. By leveraging AI and DL technologies, pathologists
Medical imaging modalities such as MRI, CT scans, can benefit from automated analysis, improved
and X-rays generate vast amounts of complex data efficiency, and the ability to handle larger volumes
that can be challenging and time-consuming for of data. DL algorithms can assist in the identification
human providers to interpret. of critical biomarkers, characterization of disease
subtypes, and prediction of disease outcomes,
The integration of AI and DL technology into ultimately leading to more personalized and precise
clinical imaging has the potential to revolutionize treatment plans.
the practice of radiology. AI can serve as a valuable
clinical decision support tool for radiologists, aiding While DL-based approaches in pathology are still in
in the interpretation of medical images and the early stages of development and require further
improving the overall delivery of care to patients. validation and integration into clinical practice, they
DL algorithms can assist in image processing tasks hold great promise for augmenting the capabilities of
by selecting and extracting relevant features from pathologists and addressing some of the challenges
the images, as well as generating new features that faced in the field. Continued research and
may enhance diagnostic accuracy. collaboration between pathologists, data scientists,
and AI experts will be crucial in realizing the full
In terms of image interpretation, DL algorithms can potential of AI-based systems in pathology.
be trained to identify and classify disease patterns
within medical images, helping radiologists detect 4. IMPACT OF AI ON DATA PRIVACY
abnormalities and suggesting appropriate care
pathways for patients. By leveraging the power of It is an important concern regarding the privacy of healthcare-
AI and ML, radiologists can benefit from advanced related data in the context of AI advancements in medicine.
image analysis techniques that augment their The increasing use of AI in healthcare does bring new
expertise and improve diagnostic accuracy. challenges to data privacy and raises the need for updated
policies and ethical standards to address these concerns. Here
The application of AI and ML in radiology has the are some key considerations:
potential to streamline workflows, reduce
interpretation errors, and improve overall patient Outdated Privacy Standards: As AI progresses, traditional
outcomes. As these technologies continue to evolve privacy standards like HIPAA may become inadequate in
and mature, they hold great promise for enhancing protecting sensitive healthcare data. The study from the
the field of radiology and transforming the way University of California, Berkeley emphasizes the need to
medical imaging is utilized in clinical practice reassess privacy standards and regulations to address the
potential risks associated with AI-enabled re-identification of
3.8. PATHOLOGY individuals, even when identifying information is removed.

Re-identification Risks: AI techniques can potentially re-

. The field of pathology faces numerous challenges,
identify individuals by learning patterns from collected data
including the increasing complexity of diseases, the
and correlating it with demographic information. This
growing amount of data in the form of biomarkers
highlights the importance of considering the privacy
and genomics information, and a shortage of
implications of daily activity tracking data from devices like
pathologists. In this context, DL-based approaches,
smartwatches and smartphones, as well as the need to develop
such as ML and CNN algorithms, can play a
robust mechanisms for de-identification and anonymization
significant role in assisting pathologists and
of healthcare data.
improving diagnostic accuracy.
Informed Consent and User Control: Users should have clear
DL algorithms can analyze large datasets, including
information about how their data will be used by AI systems
histopathological images, and extract meaningful
in healthcare. They should have the ability to provide
features and patterns that may not be easily
informed consent and the option to decline consent for certain
discernible to the human eye. This can aid in the
uses of their data. Transparency from tech companies is
classification and subtyping of diseases, leading to

© 20XX, IJCSE All Rights Reserved 6

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

crucial in ensuring that users understand and have control monitors, with AI-assisted analytics systems enables remote
over the use of their data. monitoring and management of acute conditions. This allows
for at-home healthcare and empowers physicians to make
Technical Safeguards: Tech companies developing AI better treatment decisions based on real-time patient data.
healthcare solutions should incorporate technical safeguards
into their systems to minimize the risk of data misuse. This Overall, the combination of AI and human physicians has the
includes robust data anonymization techniques, encryption, potential to revolutionize healthcare delivery by enhancing
access controls, and secure storage practices to protect precision, improving diagnostic accuracy, optimizing surgical
sensitive healthcare data throughout its lifecycle. procedures, and enabling remote monitoring and management
of patients. These advancements hold great promise for
Accountability and Regulation: There should be mechanisms improving patient outcomes and transforming the healthcare
in place to hold tech companies accountable for the protection landscape.
of healthcare data. Governments and regulatory bodies can
play a role in updating legal frameworks like HIPAA or
creating new regulations tailored to address the privacy 6. CASE STUDY: ONCOLOGISTS IN INDIA
challenges posed by AI in healthcare. Regular audits and USING IBM WATSON
compliance assessments can help ensure that data privacy
practices align with the evolving landscape of AI
technologies. IBM's Watson for Oncology is an AI-based cognitive
computing platform that has been trained by Memorial Sloan-
Addressing these challenges requires collaboration between Kettering Cancer Center to analyze data and provide
policymakers, healthcare providers, tech companies, and evidence-based treatment options for cancer patients. It is
researchers to strike a balance between the potential benefits being adopted by Manipal Hospitals in India, where there is a
of AI in medicine and safeguarding patient privacy. It is high burden of cancer cases and a shortage of oncologists.
crucial to continuously assess and update privacy policies, Watson has the ability to read and understand natural
ethical guidelines, and legal frameworks to keep pace with language and stays updated with new scientific articles and
the advancements in AI and protect patient data privacy data.
effectively.
However, it's important to note that Watson's treatment
recommendations are not based on its own insights but rather
on the training it has received from physicians at Memorial
5.  TRANSFORMING MEDICAL CARE Sloan Kettering Hospital. This can potentially limit its
WITH AI accuracy and value in different medical practices and
economic circumstances.
AI methods and systems can greatly enhance the delivery of
care when combined with human physicians. Here are the key The hope is that Watson can help physicians in Manipal
points highlighted: hospitals deliver advanced, effective, and cost-effective
treatment to their cancer patients. A study conducted by
Precision Medicine: AI and ML play a crucial role in Manipal Hospitals in 2018 showed that Watson for Oncology
enabling precision medicine by analyzing large volumes of was in agreement with the hospital's multidisciplinary tumor
clinical, genomic, and imaging data. This helps physicians board in 93 percent of breast cancer treatment decisions.
improve efficiency, enhance diagnostic accuracy, and However, it's crucial to understand that agreement with
personalize treatment based on individual patient doctors' decisions only demonstrates competency in applying
characteristics. existing methods of care, and not necessarily an improvement
upon them.
Diagnostic Efficiency: AI-based systems have the potential to
significantly improve diagnostic efficiency in various medical In summary, Watson for Oncology is an AI-based tool that
fields, including radiology, pathology, dermatology, and aims to support oncologists in making treatment decisions. Its
ophthalmology. By assisting in image interpretation and adoption in Manipal Hospitals in India can potentially address
analysis, AI can help reduce false positives, leading to the challenges posed by the increasing number of cancer
reduced patient anxiety and unnecessary procedures. cases and the shortage of specialized oncologists. However,
further research and evaluation are needed to determine its
Robotic Surgery: AI in robotic surgery can assist surgeons by true impact on improving patient outcomes and advancing
providing precision, reduced blood loss, and decreased tissue cancer care
trauma. By incorporating AI and tactile feedback, robotic
systems can enhance surgical outcomes and minimize
surgeon variations, leading to improved patient recovery. 7. IMPACT OF AI ON PHYSICIANS
Wearable Healthcare Devices: The integration of wearable The use of AI in predictive analysis and image recognition
healthcare devices, such as fitness trackers and portable has raised concerns about the potential replacement of

© 20XX, IJCSE All Rights Reserved 7

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

physicians, particularly in fields like radiology. There are

different viewpoints on this matter. One perspective suggests
that AI will become more accurate than radiologists in
interpreting complex clinical images and eventually replace Data Availability
them. Another perspective argues that AI will augment
radiologists, assisting them in routine tasks and freeing up This article is a literature review, which means that no raw
their time for more valuable activities. There is also a middle- data or code was generated from the study.
ground view that AI might replace radiologists in the distant
future, but it is not a pressing concern. Conflict of Interest

While the impact of AI on radiologists is difficult to predict, The authors did not have any commercial or financial
it seems more reasonable to expect that AI will augment relationships that could be seen as a possible conflict of
physicians rather than replace them. There are several interest while conducting the research.
limitations to AI that support this conclusion. AI cannot
engage in high-level conversations or establish the same level Funding Source
of trust, reassurance, and empathy as human doctors, which
are crucial aspects of the doctor-patient relationship. The authors did not receive any funding for this work.
Physicians are still necessary for physical examinations and
fields like neurology that require direct interaction with Authors’ Contributions
patients and critical thinking. Additionally, AI-based systems
rely on precedents and may struggle with novel or unusual All authors have contributed significantly and intellectually to
cases where there is no prior example to draw from. the work.
Suhana Sapui analyzed and prepared the rough draft and
AI-based systems are likely to support physicians by assisting Supriya Ghosh reviewed the informations and prepared the
them in routine tasks and improving efficiency. However, the final draft.
role of physicians in high-level decision-making, patient
interaction, and handling complex cases will remain crucial. Acknowledgements
The traditional physician-patient relationship is unlikely to be
replaced entirely by AI, as the skills and expertise of I would like to express my deep and sincere gratitude to my
physicians are essential in providing comprehensive and research supervisor, Dr. Radhakrishna Jana, Professor of JIS
personalized healthcare. University, for giving me the opportunity to do the research
and providing invaluable guidance throughout this research.
8. CONCLUSION His dynamism, vision, sincerity and motivation have deeply
inspired me. He has taught us the methodology to carry out
As precision and personalized medicine continue to evolve, the research and to present the research works as clearly as
the reliance on data, including clinical, genomic, and imaging possible. It was a great privilege and honor to work under his
data, is expected to increase. The integration of AI guidance.
technology with medicine is becoming more prominent, as it
has the ability to analyze and make sense of vast amounts of
data. This trend is being recognized by medical schools, References
which are incorporating emerging technology curricula into
their programs. Courses on technology infrastructure, 1. Abhimanyu S Ahuja (2019) The impact of artificial
machine learning (ML), deep learning (DL), and data intelligence in medicine on the future role of the
management are being offered alongside traditional biology physician
classes.
2. Giovanni Briganti and Olivier Le Moine (2019-20)
AI will play a supportive role in medicine by assisting in the Artificial Intelligence in Medicine: Today and
analysis of diverse data sources that patients and healthcare Tomorrow
institutions generate continuously. This will enable
physicians to allocate more of their valuable time to direct 3. Amisha, Paras Malik, Monika Pathania, Vyas
patient care, enhancing the human aspect of medicine. While Kumar Rathaur (2019) Overview of artificial
it is unlikely that AI will replace physicians in the foreseeable intelligence in medicine
future, it is crucial for medical professionals to familiarize
themselves with the fundamentals of AI technology and 4. Md Mobashir Hasan Shandhi and Jessilyn P.
understand how AI-based solutions can improve patient Dunn (2022) AI in medicine: Where are we now and
outcomes. In the future, physicians who embrace and where are we going?
effectively utilize AI may outperform those who do not adapt
to these technological advancements. 5. Topol EJ (2019) A decade of digital medicine
innovation

© 20XX, IJCSE All Rights Reserved 8

International Journal of Computer Sciences and Engineering Vol.X(X), Dec20XX

6. Shah NR (2019) Health care in 2030: will artificial

intelligence replace physicians?

7. Cavallo J. (2017) How Watson for oncology is

advancing personalized patient care

8. Kent J. (2018) How artificial intelligence is

changing radiology, pathology

9. Venkatesan M. (2018) Artificial intelligence vs.

Machine learning vs. Deep learning

10. Society for the Study of Artificial Intelligence and

Simulation of Behaviour (2018) What is artificial
intelligence

11. Hamlet P, Tremblay J (2017) Artificial Intelligence

in medicine

12. Mintz Y, Brodie R (2019) Introduction to artificial

intelligence in medicine

AUTHORS PROFILE
Suhana Sapui has completed her ICSE and
ISC from Delhi Public School Megacity,
Kolkata and is currently pursuing Bachelors of
Technology (B.Tech) degree from JIS
University

Supriya Ghosh has completed his WBBSE

and WBCHSE from The Scottish Church
Collegiate School, Kolkata and is currently
pursuing Bachelors of Technology (B.Tech)
degree from JIS University

© 20XX, IJCSE All Rights Reserved 9