Artificial Intelligence in Medical Diagnosis: A Human-Centred Approach

---

Anjali Agrawal¹ , Rajni Yadav²

¹Amity Business School, Amity University Chhattisgarh, India, 493225

²Amity Institute of Pharmacy, Amity University Chhattisgarh, India, 493225

Corresponding Author Email: ryadav@rpr.amity.edu

DOI : https://doi.org/10.51470/APR.2026.05.01.28

Abstract

Keywords

Download this article as:

---

1. INTRODUCTION

Though technology comprises various facets, one broad definition is technology as the use of computers, computer systems, and other forms of technology to manage, process, and disseminate information. Technology covers a wide range of specialities. For instance, cybersecurity and IT Support are included as part of technology. A business firm can hardly function without technology today. Concerns addressed by business technology include data storage, communication, software development, network administration, and system security^.[1][2]

The term artificial intelligence (AI) was coined at the Dartmouth Summer Workshop in 1956, when AI was generally described as “thinking machines.” Simply, AI means a machine’s ability to learn and recognise patterns and relations from a large mass of data and then use what it has learned in making decisions when presented with new, unfamiliar information.

Disease diagnosis involves recognising a disease, disorder, or any other condition that an individual may possess based on signs and symptoms. Extensive data sets exist; however, there is a scarcity of tools capable of accurately identifying patterns and generating predictions. The conventional techniques employed to identify a disease are hands-on and susceptible to mistakes. AI offers numerous benefits in healthcare, but it also has its disadvantages. The use of AI in a responsible and ethical way is a key concern. When handling sensitive medical information, it is crucial to ensure patient confidentiality and data protection.^[3]

Although AI in X-ray and CT imaging mainly utilises large standardised image collections for automatic pathology detection and diagnosis, the fixed characteristics of these imaging methods limit AI’s effectiveness in real-time diagnostics. Magnetic Resonance Imaging (MRI) provides significant analytical opportunities for AI due to its superior resolution and tissue differentiation; however, its expense, lengthy scanning periods, and intricate image reconstruction methods have steered AI initiatives towards improving image quality and shortening scan times. In contrast, ultrasound, being a dynamic and real-time imaging technique, offers unique benefits.^[1][2]

Artificial intelligence has the potential to revolutionise healthcare diagnosis, treatment, and prediction. How does this technology work? This paper presents a solution that will use artificial intelligence to improve healthcare outcomes. The integration of AI into healthcare operations can result in better treatment plans, more accurate patient assessments, and improved diagnoses. Ethical and privacy concerns, as well as efficient deployment methodologies, must be addressed in order for AI to reach its full potential in healthcare.

1.1 Design and Implementation

Artificial intelligence has the potential to revolutionise healthcare diagnosis, treatment, and prediction. How does this technology work? AI is being utilised to improve healthcare outcomes, and the solution described in this paper is currently under development. AI can be used to optimise treatment plans, predict patient outcomes and improve diagnosis in healthcare. Why is this so? Despite its potential, AI in healthcare poses ethical challenges, as does privacy, and requires efficient deployment methods.

• ARTIFICAL INTELIIGENCE AND DIAGNOSIS

Accurate and timely diagnosis is a significant hurdle in healthcare. AI-driven technologies have been proven to be highly efficient in aiding medical professionals in diagnosing a wide range of illnesses. These systems process and analyse vast amounts of medical data, such as patient records, imaging results, and genetic data. Through the use of machine learning algorithms and deep learning, these systems can identify patterns in information, identify abnormalities, or provide diagnostic information that can be used for further improvement. Many different areas have seen the effective use of AI-based diagnostic tools, such as computer-aided detection (CAD) and computer–aided diagnosis (CADx) systems. Images from X-rays, CT scans and MRI images are processed by AI algorithms to identify anomalies in imaging techniques used in radiology to aid doctors in making more accurate diagnoses.

Why? Similarly, pathologists have utilised AI algorithms in pathology to assist them in analysing tissue samples and identifying malignant cells.

Several processes are involved in the application of AI for diagnosis. First, there is a significant need to collect and store many medical data in pristine condition. Why? The data is utilised to instruct AI systems on the use of supervised learning, unsupervised learning, or reinforcement learning. How does this work? These improved models are further validated and enhanced by the use of new datasets. Healthcare practitioners can use AI models to provide real-time insights and support, which can improve diagnostic accuracy and patient outcomes.

MEDICAL DIAGNOSIS PROCESS USING AI ML

A paradigm shift that we once deemed impossible is now brought about by machine learning. This branch of Artificial Intelligence is a specialisation that allows systems to learn patterns in data and improve their performance on their own, moving away from complete manual programming or human judgment.

Data collection begins with the gathering of data. A strong data foundation is necessary for any ML system. The system’s performance is directly influenced by the volume and integrity of the information collected. Essentially, the purpose of this stage is to gather in-depth knowledge about a particular audience or subject variables.^[4]

After gathering the data, it is subjected to preprocessing. Data cleansing involves converting unrefined or dispersed data into a manageable format that is suitable for analysis. The information is refined to make precise decisions.

To model, it is necessary to choose an algorithm that works best for the specific task being executed.

In this stage, a particular mathematical model is chosen in response to the nature of the problem. To enhance accuracy and predictive power, Supervised Learning is used for most predictive tasks, where the model is trained on labelled examples to improve its learning process.

It is necessary to evaluate the model using a particular set of numbers.

Specific benchmarks and desired outcomes are used to assess the model’s performance after training. To ensure a measurable boost in performance, it’s essential to compare these results with previous versions or baseline data. This step can include the parameters.

To achieve optimal performance, the model must undergo “fine-tuning,” which involves modifying internal settings like the number of training iterations or steps to ensure the algorithm follows the most efficient path to a solution. To test the model against actual data. A final analysis involves utilising the model to forecast outcomes from an unfamiliar “test” dataset. The model is deemed reliable and suitable for practical forecasting when it aligns with the findings of domain experts or established facts.

ML is utilised for the purpose of detecting diseases, as explained below.

Information Gathering: Compiling specific medical examination findings and patient manuals.

Utilising feature extraction to identify and separate the biological markers or attributes that are most effective in predicting disease.

Targeting: Sorting the data to identify the most relevant factors for a conclusive analysis.

The use of multiple classification methods to validate the correctness of the processed data and enhance the precision of disease diagnosis through Model Evaluation.^[14][9]

Modern healthcare is being transformed by the integration of AI and Machine Learning, which provide predictive insights and automated image analysis to catch diseases like cancer and Alzheimer’s earlier than ever before. This is supported by advanced imaging (MRI/CT/PET) and genomic diagnostics that identify biological markers at a molecular level, allowing for highly personalised treatments. Simultaneously, the rise of Wearable Technology (IoMT) and Telemedicine has shifted care from the hospital to the home, enabling continuous real-time monitoring of vital signs and expanding access to specialist expertise regardless of geography.

• BENEFITS OF AI IN MEDICAL DIAGNOSIS
• Improved Precision and Early Identification 

AI algorithms can process thousands of images or data points much faster than humans and with great accuracy. For instance, in radiology, AI-driven tools have demonstrated the ability to detect cancers in mammograms or lung nodules in CT scans earlier than conventional techniques. Detecting diseases early enhances the likelihood of effective treatment and survival.^[14]

• Minimising Human Mistakes 

Diagnostic errors by humans may result from fatigue, oversight, or cognitive biases. AI systems offer a second opinion, serving as a safeguard to reduce misdiagnoses. They can highlight inconsistencies or unusual findings that need further examination, thereby supporting physicians’ decisions rather than replacing them.

• Greater Efficiency and Cost Savings 

By automating labour-intensive tasks such as image analysis or reviewing patient records, AI allows healthcare providers to dedicate more time to direct patient care. This increased efficiency can reduce healthcare expenses and help mitigate shortages of specialised medical professionals, particularly in underserved regions.

• Customised Diagnosis and Treatment 

AI can combine extensive data, including genetic and lifestyle information, to customise diagnoses and treatment plans for each individual. Personalised medicine results in more targeted therapies and fewer adverse effects.

Illustrations of AI Use in Medical Diagnosis 

1. Radiology: AI assists radiologists in interpreting X-rays, MRIs, and CT scans to identify fractures, tumours, or infections^[7][11][8]. 

2. Dermatology: Image recognition technology aids in detecting skin cancers and other dermatological conditions from photographs. 

3. Pathology: AI evaluates tissue biopsies for cancerous changes or infections with higher throughput. 

4. Ophthalmology: Automated retinal imaging supports the diagnosis of diabetic retinopathy and other vision-threatening diseases. 

5. Cardiology: AI analyses ECGs and forecasts risks of heart attacks or arrhythmias.^[5]

• IMPEDIMENTS AND NUANCES

Although the promising potential exists, AI in medical diagnosis encounters various obstacles:

1. Data Integrity and Prejudice
   AI models need extensive and varied datasets for training. When the data is skewed or lacking, the AI’s precision declines and can reinforce health inequities.
2. Explainabilty

Numerous AI systems, particularly deep learning models, function as black boxes where their decision-making process remains opaque. The absence of explainability leads to trust and ethical issues.

• Oversight and Responsibility

Establishing responsibility for AI mistakes presents a complicated legal and moral challenge. Regulatory structures continue to develop.

• Integration

Integrating AI into current clinical processing without causing interruptions requires thoughtful planning and education.^[16]

• THE FUTURE OF INTEGRATION OF ARTIFICIAL INTELLIGENCE IN CLINICAL DECISION SUPPORT

AI’s role as a diagnostic tool will only continue growing in the future. As natural language processing, computer vision, and data integration continue to improve, AI’s ability to perform diagnoses will continue to expand.^[9] Hybrid models using AI in conjunction with clinicians’ expertise will soon become the norm; this will enable greater precision in medicine. In addition, AI diagnostic tools could provide access to more expert-level interpreters for patients in remote areas or areas lacking patient care by utilising mobile devices. Research is also being conducted to determine how AI can be utilised for predicting disease outbreaks or tracking trends in the health of the general public.

• CONCLUSION

The use of AI in healthcare provides beneficial results by improving the diagnosis process and detecting diseases at early stages, which helps in selecting appropriate treatment plans. Another important point is that we examined three AI techniques—fuzzy logic, machine learning, and deep learning—that are commonly used in healthcare, and we generated our results using these methods. We also analysed the effect of each AI technique based on the frequency of influence recorded in the papers. The major medical areas we reviewed included cardiology, neurology, cancer, kidney disease, diabetes, cholera, and dental disease, all using AI diagnostic criteria. Furthermore, we discovered significant differences in the papers depending on the type of disease. In this study, we observed that AI is not limited to identifying specific diseases. We can use different AI techniques to detect any type of disease or to improve the diagnosis process for all diseases. Therefore, we believe this survey will be useful for future research. Additionally, in this paper, we noted that over 91% of AI methods reported a positive impact on disease diagnosis. The ability of AI to detect diseases should not be overlooked. People are increasingly finding out the value and convenience of cloud computing and are accessing this type of service more and more. Medical science in technology. Medical science in technology (Medical Technology) is the application of scientific knowledge, engineering, and innovation to create tools, devices, and solutions that improve health, enhance patient care, and increase efficiency in screening, diagnosis, treatment, and rehabilitation.

• REFERENCES
• Kaur, S., Singla, J., Nkenyereye, L., Jha, S., Prashar, D., Joshi, G. P., … & Islam, S. R. (2020). Medical diagnostic systems using artificial intelligence (AI) algorithms: principles and perspectives. Ieee Access, 8, 228049-228069.
• Tian, M., Shen, Z., Wu, X., Wei, K., & Liu, Y. (2023). The application of artificial intelligence in medical diagnostics: A new frontier. Academic Journal of Science and Technology, 8(2), 57-61.
• Fetzer, J. H. (1990). What is artificial intelligence?. In Artificial intelligence: Its scope and limits (pp. 3-27). Dordrecht: Springer Netherlands.
• Szolovits, P., Patil, R. S., & Schwartz, W. B. (1988). Artificial intelligence in medical diagnosis. Annals of internal medicine, 108(1), 80-87.
• Dilsizian, S. E., & Siegel, E. L. (2014). Artificial intelligence in medicine and cardiac imaging: harnessing big data and advanced computing to provide personalized medical diagnosis and treatment. Current cardiology reports, 16(1), 441.
• Göndöcs, D., & Dörfler, V. (2024). AI in medical diagnosis: AI prediction & human judgment. Artificial Intelligence in Medicine, 149, 102769.
• Al-Antari, M. A. (2023). Artificial intelligence for medical diagnostics—Existing and future AI technology!. Diagnostics, 13(4), 688.
• Park, S. H., Han, K., Jang, H. Y., Park, J. E., Lee, J. G., Kim, D. W., & Choi, J. (2023). Methods for clinical evaluation of artificial intelligence algorithms for medical diagnosis. Radiology, 306(1), 20-31.
• Das, S., Biswas, S., Paul, A., & Dey, A. (2017, July). AI Doctor: An intelligent approach for medical diagnosis. In Industry Interactive Innovations in Science, Engineering and Technology: Proceedings of the International Conference, I3SET 2016 (pp. 173-183). Singapore: Springer Singapore.
• Fan, W., Liu, J., Zhu, S., & Pardalos, P. M. (2020). Investigating the impacting factors for the healthcare professionals to adopt artificial intelligence-based medical diagnosis support system (AIMDSS). Annals of Operations Research, 294(1), 567-592.
• Srivastav, S., Chandrakar, R., Gupta, S., Babhulkar, V., Agrawal, S., Jaiswal, A., … & Wanjari, M. (2023). ChatGPT in radiology: the advantages and limitations of artificial intelligence for medical imaging diagnosis. Cureus, 15(7).
• https://www.researchgate.net/publication/364826401_Understanding_basic_principles_of_Artificial_Intelligence_a_practical_guide_for_intensivists
• Deshmukh, A. (2024). Artificial Intelligence in Medical Imaging: Applications of Deep Learning for Disease Detection. Universal Research Reports
• Bhatt, J., Jain, S., & Bhatia, D. D. (2025). Artificial intelligence in healthcare diagnosis: Evidence-based advances. Sensors & Diagnostics.
• Ahsan, M. M., Akter Luna, S., & Siddique, Z. (2022). Machine-Learning-Based Disease Diagnosis: A Comprehensive Review. Healthcare
• Gerke, S., Minssen, T., & Cohen, G. (2020). Ethical and legal challenges of artificial intelligence-driven healthcare. Artificial Intelligence in Healthcare, 295–336.