Advancing the Future of Pathology with Artificial Intelligence

Aneesh Sathe, CEO & Co-Founder, Qritive

Pathology plays an integral role in clinical decisions by enhancing the objectivity of patient diagnosis and disease care. However, it continues to face challenges around manpower and resource constraints and increasing complexity of pathology analysis. This article explores some of these challenges and the potential of AI and digital adoption in advancing the future of pathology.

70 per cent of all clinical decisions involve pathology investigation1. As such, pathology plays an integral role in our healthcare systems, as decisions made around a patient’s diagnosis and treatment are dependent on pathology-based data and analysis.3

Yet, despite its prevalence, pathology – much like other medical fields – continues to face its share of challenges around manpower and resource strains, which have been exacerbated by COVID-19. This has sparked discussions on how we could possibly navigate and shape the future of pathology moving forward, by exploring new innovations and solutions in an increasingly technologically driven world.

Challenges in Pathology
While COVID-19 has served as a major catalyst for the profound changes happening across our healthcare systems, Asia has long been confronted with other pressing issues such as a growing geriatric population and the rise of chronic diseases.

Home to 60 per cent of the world’s population, Asia accounts for half the burden of cancer worldwide, and the incidence of cancer cases is expected to rise from 6.1 million in 2008 to 10.6 million in 2030 across the region.2 Considering this, and the backlog of healthcare procedures accumulated due to COVID-19 lockdowns, the pressure for pathologists to meet this demand is mounting. Now more than ever, pathologists and their expertise in disease study is paramount in informing more accurate and confident diagnosis for primary care physicians to meet patient throughputs.

However, while the demand for healthcare screening is rising exponentially, the world faces a shortage of pathologists who can meet this demand. In the US alone, the number of pathologists dropped by 17.5 per cent within a decade, from 2007 to 2017.3 Without sufficient pathologists to conduct laboratory testing and analyse clinical abnormalities that serve as markers of diseases, this could result in longer waiting times and delayed diagnosis for patients. Ultimately, it reduces the chances of survival for patients with chronic diseases and exacerbates the disease burden across the region. The consequence of having a shortage of pathologists as such, is two-fold.

Additionally, pathology analysis has grown increasingly complex over the years as better treatments are available and picking the right treatment requires deeper analysis. Accompanied by the lack of manpower, this could result in more subjectivities in data analysis and diagnosis, adding on to the stress that pathologists are already experiencing due to the high volumes of workload.

Rise of Digital Pathology
In light of these challenges, digital adoption is key in transforming traditionally analog industries like pathology. In the last three years, COVID-19 has catapulted digital transformation, and this has contributed to the rise of digital pathology-based approaches such as whole slide imaging (WSI) and AI-driven solutions. These solutions could now substitute the use of traditional light microscopes,4 allowing pathologists to analyse information that is beyond the human ability to do so.5 While other fields of medicine like radiology and cardiology have had the option of incorporating AI in their workflows much earlier, digital adoption in pathology is still in its nascent stage, albeit a promising one. By 2027, Asia’s digital pathology market is projected to reach US$125 million, from US$74 million in 2022.6

Broadly defined as virtual microscopy where digital information is being captured, managed, analysed, and interpreted from a glass slide,7 digital pathology is crucial in helping pathologists to facilitate remote diagnostic work, image analysis, clinical trial reviews, collaborations, and teleconsultation.5 This helps to improve overall workflow efficiency at a time when demand for screenings is mounting, and there is a shortage of manpower. With the advancement of WSI, more data can now be used to train Artificial Neural Network (ANN) models which can manage and integrate complex data effectively to improve diagnosis, classification and prediction of diseases, and maximise patient care.8 The adoption of AI in digital pathology is also helpful in automating time-consuming diagnostic tasks such as the counting of mitoses and screening of cancer types;9 and in simplifying complex tasks including triaging biopsies, enabling pathologists to focus more time on providing patient-centric care.

Numerous studies have demonstrated the utility of AI in improving efficiency and accuracy. For instance, AI-based modules, when used in quantitative immunohistochemistry, has shown to increase the accuracy of output and reduce discordance between pathologists by 82 per cent. As the estimation of the Ki-67 proliferation index is traditionally a manual and laborious task, AI makes pathologists’ work easier by improving the efficiency and accuracy of scoring,10 which is key to the early detection and diagnosis of Sarcomas – rare cancers that develop in our bones and soft tissues. In another study, it was found that AI is capable of picking up high-risk colorectal cancer features and could serve as an effective screening tool by outlining malignant glands.10 This helps in freeing up pathologists’ time for more important tasks, and in tackling the growing colorectal cancer burden in Asia.

AI modules could also remedy differences in Gleason grading amongst pathologists and reduce concurrences in manual measurements by identifying Gleason patterns and quantifying tumour tissues accurately.

Clearly, the synergy of AI and digital pathology has opened up a world of image-based possibilities that could transform disease diagnosis and optimise patient outcomes.8 The integration of AI and Machine Learning (ML) techniques in digital tissue images has helped in detecting, quantifying and characterising specific tissue structures and novel biomarkers for precision medicine and disease management.10 In the study of breast cancer for instance, it was found that deep neural networks can generate scores that differentiate between low- and high- grade tumours and could evaluate tissue samples that would generate a score showcasing the probability of survival for patients.6 This could prove beneficial in enhancing the precision and capabilities of digital pathology moving forward. The adoption of digital pathology and AI will certainly be a key driver in pushing pathology to new frontiers.

Considerations on AI
Nevertheless, while AI brings expansive value to the field of pathology, considerations around its applicability and practicality in real-life remain a key concern.

Applicability in Real-Life: Resources & Financing
While many papers pertaining to the study and application of AI in pathology have been published, only a handful were manifested into commercial products for clinical use.10 Aside from regulatory challenges, the process of translating and adapting new discoveries to suit real-life applications and environments require substantial time and effort, and this is a gap that is often overlooked when it comes to the adoption of AI.11 Even in instances when this might be possible, barriers such as having inadequate financing and technological infrastructure to support these new innovations remain. For instance, financing is a key consideration when it comes to the development of AI solutions, as personnel and procurement of sufficient amounts of training and testing data are required. To curb these high costs, hiring employees who are skilled and knowledgeable in the field of AI and pathology is paramount.10 For AI to be truly integrated into the norm, pathology labs must be supported and equipped with the latest digital pathology systems to integrate new solutions.10

To do so, key stakeholders including government and hospitals may consider prioritising or providing more investment in accelerating digital transformation. There is a crucial need to drive understanding of the importance of digital adoption in strengthening and creating more positive health outcomes for both pathologists and patients.

Supporting Pathologists
Another challenge when it comes to the adoption of AI is managing pathologists’ perceptions and cultivating their understanding of the benefits it offers. Given that innovations in AI are constantly evolving and the field of medicine is always advancing, pathologists may find it challenging to keep up to date with the current trends. While numerous validation studies have been published around the use of AI, pathologists remain concerned with the ease of use of these devices, and its reliability in delivering the results it promises. As such, as pathologists are the primary users of AI, it is vital to support them by bridging these knowledge gaps and educating them on how AI can be integrated into their daily workflows to improve efficiency and accuracy. One of the ways to do so is to involve pathologists from the early stages of developing these solutions, giving them ample time to try out new prototypes. It is also vital for industry stakeholders to organise educational webinars to facilitate open discussions on new innovations.

The Way Forward
Given the speed at which digital transformation is evolving across Asia, there is no doubt that digital adoption will continue to advance the future of pathology. However, there have been concerns around how AI may eventually replace pathologists.5 As such, it is crucial to find a balance between adopting technology and maintaining the human touch in healthcare. New technologies should serve as tools that can assist pathologists in their tasks, enhancing the quality and accuracy of diagnosis and patient care.

In this piece, we have highlighted some of the benefits that digital pathology and AI offer in addressing the pain points faced by pathologists in this evolving environment. While the integration of new innovations is not a straightforward process, as we have outlined under limitations, there is optimism in how they could be increasingly incorporated into the field of pathology. What is clear is that collaboration between different stakeholders is key to accelerating this transformation.


1. Badrick, Tony. “Evidence-based laboratory medicine.” The Clinical biochemist. Reviews vol. 34,2 (2013): 43-6.
2. Sankaranarayanan, Rengaswamy et al. “Managing the changing burden of cancer in Asia.” BMC medicine vol. 12 3. 8 Jan. 2014, doi:10.1186/1741-7015-12-3
3. Metter DM, Colgan TJ, Leung ST, Timmons CF, Park JY. Trends in the US and Canadian Pathologist Workforces From 2007 to 2017. JAMA Netw Open. 2019;2(5):e194337. doi:10.1001/jamanetworkopen.2019.4337
4. Tizhoosh, Hamid Reza, and Liron Pantanowitz. “Artificial Intelligence and Digital Pathology: Challenges and Opportunities.” Journal of pathology informatics vol. 9 38. 14 Nov. 2018, doi:10.4103/jpi.jpi_53_18
5. Baxi, V., Edwards, R., Montalto, M. et al. Digital pathology and artificial intelligence in translational medicine and clinical practice. Mod Pathol 35, 23–32 (2022).
6. Market Data Forecast, “Asia Pacific Digital Pathology Market Research Report”, January 2022,
7. Niazi MKK, Parwani AV, Gurcan MN. Digital pathology and artificial intelligence. Lancet Oncol. 2019 May;20(5):e253-e261. doi: 10.1016/S1470-2045(19)30154-8. PMID: 31044723; PMCID: PMC8711251.
8. Cui, M., Zhang, D.Y. Artificial intelligence and computational pathology. Lab Invest 101, 412–422 (2021).
9. Homeyer, André et al. “Artificial Intelligence in Pathology: From Prototype to Product.” Journal of pathology informatics vol. 12 13. 22 Mar. 2021, doi:10.4103/jpi.jpi_84_20
10. USCAP 2022 Abstracts: Informatics (977-1017). Mod Pathol 35, 1163–1210 (2022).
11. Rakha EA, Toss M, Shiino S, Gamble P, Jaroensri R, Mermel CH, Chen PC. Current and future applications of artificial intelligence in pathology: a clinical perspective. J Clin Pathol. 2021 Jul;74(7):409-414. doi: 10.1136/jclinpath-2020-206908

Aneesh Sathe

Aneesh Sathe is the CEO and Co-Founder of Qritive. With 13 years of experience in the biotechnology industry, Aneesh leads a team that shares and expands his vision of making a large impact by bringing advanced computational tools to improve healthcare. From being a PhD student at NUS, Aneesh’s entrepreneurial journey is defined by a desire to play with technology and help society with new discoveries and innovation.

Latest Issue
Get instant
access to our latest e-book
Sanofi Harvard Medical School - Leadership in Medicine: Southeast Asia Medlab Middle East 2023 Dubai Derma 2023