Indian School of Business

ISB368

March 28, 2023

Vijaya Sunder M | Rithica Mamidi

AN ARTIFICIAL INTELLIGENCE (AI)-BASED DISRUPTIVE INNOVATION IN

CATARACT SCREENING: THE CASE OF e-PAARVAI
INTRODUCTION

It was a summer morning in early 2022. A local newspaper in Tamil Nadu reported: “The e-Paarvai app
is poised to help the 20 million residents of Tamil Nadu aged above 50, who are the most susceptible
to cataract. However, this solution can just as well be implemented across states, reducing the burden
on the healthcare system and reliance on the limited numbers of ophthalmologists.” Azhagu Kanmani
was overwhelmed with joy upon reading this news about the potential scale-up of e-Paarvai beyond
Tamil Nadu.

e-Paarvai, an artificial intelligence (AI)-based application, was initially launched in February 2021 to
detect cataract in the south Indian state of Tamil Nadu. It came into being due to the collaborative
efforts of the Tamil Nadu e-Governance Agency (TNeGA) and the Tamil Nadu State Blindness Control
Society (TNSBCS), which aimed to eradicate cataract from the state. Kanmani, a 63-year-old resident
of Nagapattinam district of Tamil Nadu, was among the first few who underwent surgery after her
cataract was detected during the e-Paarvai pilot in January 2021.

e-Paarvai was conferred the NASSCOM1 AI Gamechangers Award in December 2021 for being among
the best AI-based innovations that created an immense social impact (see Exhibit 1). The application
had helped screen about 25,000 cataract patients in more than 30 districts in rural Tamil Nadu within
ten months of its launch. The solution was further scaled to the entire state in early 2022, and could
operate even in locations with low Internet connectivity, such as remote rural and hilly areas.

Thrilled to read the success story of e-Paarvai, Kanmani connected with Krishnamurthy Govindarajan,
her cousin, over the phone. They recollected speaking about the e-Paarvai scale-up in one of their
previous meetings and were happy to see it becoming a reality now.

1 National Association of Software and Service Companies, popularly known as NASSCOM is an Indian non-governmental
trade association and advocacy group, focused mainly on the technology industry of India
Professor Vijaya Sunder M and Rithica Mamidi prepared this case solely as a basis for class discussion. This case is not
intended to serve as an endorsement, a source of primary data, or an illustration of effective or ineffective management.
This research is funded by Center for Business Innovation, Indian School of Business. This case was developed under the aegis
of the Centre for Learning and Management Practice, ISB.

Copyright @ 2023 Indian School of Business. The publication may not be digitised, photocopied, or otherwise reproduced,
posted or transmitted, without the permission of the Indian School of Business.

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
CATARACT: A LEADING CAUSE OF VISION IMPAIRMENT

Chronic eye disease has been a significant health problem globally. In 2020, 1.1 billion people lived
with vision loss worldwide, of which over 43 million were blind.2 An estimated 295 million people had
moderate or severe vision impairment, 257 million had mild visual impairment, and 507 million
suffered from other associated uncorrected visual impairment (see Exhibit 2 for the World Health
Organization [WHO] Definitions of Blindness and Visual Impairment).

Despite awareness of the prevalent eye diseases, early detection and treatment of the associated
chronic problems had been a low-focus area in many communities, leading to permanent eye damage.
In older adults, vision impairment led to social separation, difficulty in walking, a greater risk of falls
and fractures, and the likelihood of an early entry into nursing or care homes. Research showed that
the prevalence of vision loss increased rapidly with age, and most patients were over the age of 60
years.3 With the aging population estimated to triple from 140 million in 2019 to 420 million in 2050,4
chronic eye conditions could also increase. Consequently, in line with the UN Sustainable
Development Goal 3, the WHO proposed many people-centered eye care programs to address this
global issue. Although several governments had taken initiatives to address eye disorders, their
incidence continued unabated in 2022.

Cataract was the leading cause of needless and frequently avoidable chronic eye disorders. As of 2020,
around the world, 17 million people became blind due to cataract, and 83 million experienced vision
impairment and sight restoration with cataract surgery. 5 About 50% of all blindness in low- and
middle-income countries was attributed to cataract,6 and in India, this percentage rose to 66.2%.7

Although cataract was relatively simple to treat and remove surgically, access to the required eye care
resources was extremely limited in emerging economies. As of 2021, in many developing countries,8
about 99% of people with cataract resided in rural or semi-urban communities that lacked adequate
health infrastructure. If not detected and treated within the stipulated period, cataract leads to
blindness. Timely cataract removals are associated with positive socioeconomic effects. For example,
an Indian study showed an increase in economic productivity of up to 1,500% of the surgery cost

2 IAPB (International Agency for the Prevention of Blindness).(n.d.). VISION 2020. Retrieved February 20, 2023, from The
International Agency for the Prevention of Blindness website: [Link]
3 Hashemi, H., Pakzad, R., Yekta, A., Aghamirsalim, M., Pakbin, M., Ramin, S., & Khabazkhoob, M. (2020). Global and

regional prevalence of age-related cataract: A comprehensive systematic review and meta-analysis. Eye, 34(8), 1357–1370.
[Link]
4 IAPB (International Agency for the Prevention of Blindness). (2020). Projected change in vision loss 2020 to 2050.

Retrieved February 20, 2023, from The International Agency for the Prevention of Blindness website:
[Link]
5 IAPB (International Agency for the Prevention of Blindness). (2020). Causes of vision loss. Retrieved February 20, 2023,

from The International Agency for the Prevention of Blindness website: [Link]
of-vision-loss/
6 Kohli, T. (2021, June 28). To end extreme poverty, we must also end blindness. Retrieved February 20, 2023, from World

Economic Forum website: [Link]

blindness/
7 NPCB & VI (National Programme for Control of Blindness and Visual Impairment). (2019). National Blindness and Visual

Impairment Survey India (2015-2019) – A summary report. Retrieved from NPCB&VI website:
[Link]
8 SEE International. (n.d.). Cataracts. Retrieved February 20, 2023, from SEE International website:

[Link]

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 2

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
during the first postoperative year. 9 However, several factors hampered the timely screening and
detection of cataract in rural areas.

First, rural communities lacked awareness of eye health and ophthalmologic diseases, and this was
exacerbated by indifference. One reason for this could be the lower literacy rate. Second, the lack of
ophthalmologists and their unequal distribution was a significant problem. The incidence of
ophthalmologists per million population was about 4 in low-income countries in 2021, compared with
a mean of 76.2 in high-income countries.10 The registered ophthalmologist-to-population ratio in India
stood at 1:250,000 in rural areas,11 against the WHO standard of 1:50,000 (in developing economies)
and 1:10,000 (in developed economies).12

Third, in India, many daily wage laborers naturally preferred not to lose income by walking long
distances to get their eyes checked. The affordability of eyecare added to the complexity of the
problem, as most rural residents (about 85.9%) had no health insurance coverage.13 To address the
affordability problem to some extent, the Government of India, under their flagship program
Ayushman Bharat, provided a framework for equitable health services covering a sum of INR 0.5
million per family per year.14

Lastly, many rural residents had the mistaken notion that cataract surgery was an expensive, premium
health service and was therefore out of bounds for the lower-income group.

THE STATE OF CATARACT IN TAMIL NADU: THE ROLE OF THE TNSBCS

As in other Indian states, untreated cataract was one of the primary reasons for blindness in Tamil
Nadu in 2020. With about 0.19 million cataract patients,15 the prevalence of blindness in the state was
4 per 1,00016 against the national average of 3.6 per 1,000 population, based on a National Blindness
and Visual Impairment 2015–2019 Survey (see Exhibit 3).17 These numbers were significantly high in
the rural communities of Tamil Nadu, which constituted 51.5% of the state’s population (about 37.2

9 Unite for Sight. (2021). Module 12: Poverty and Blindness.

[Link]
10 Resnikoff, S., Lansingh, V. C., Washburn, L., Felch, W., Gauthier, T.-M., Taylor, H. R., Eckert, K., Parke, D., & Wiedemann,

P. (2019). Estimated number of ophthalmologists worldwide (International Council of Ophthalmology update): Will we
meet the needs? British Journal of Ophthalmology, 104(4), 588–592. [Link]
11 NASSCOM (National Association of Software and Service Companies). (2021). AI gamechangers: Accelerating India with

innovation. New Delhi: NASSCOM. [Link]

12 Naidoo, K., Govender-Poonsamy, P., Morjaria, P., Block, S. S., Chan, V. F., Yong, A. C., & Bilotto, L. (2022). Global mapping

of optometry workforce. Research Square preprint. [Link]

13 Yadavar, S. (2019, December 15). 90% poorest have no health insurance, reel under high medical costs. Business

Standard. Retrieved from [Link]

insurance-reel-under-high-medical-costs-report-119121100371_1.html.
14 PIB (Press Information Bureau). (2021). Ayushman Bharat Pradhan Mantri Jan Arogya Yojana (AB-PMJAY). Press release.

Ministry of Health and Family Welfare. Retrieved from [Link]

15 NPCB & VI (National Programme for Control of Blindness and Visual Impairment). (n.d.). NPCBVI dashboard. Retrieved

August 12, 2022, from [Link]

16 Health & Family Welfare Department, Government of Tamil Nadu. (n.d.). Tamil Nadu State Blindness Control Society.

Retrieved February 20, 2023, from: [Link]

17 NPCB & VI (National Programme for Control of Blindness and Visual Impairment). (2019). National Blindness and Visual

Impairment Survey India (2015-2019) – A summary report. New Delhi: NPCB & VI. Retrieved from NPCB & VI website:
[Link]

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 3

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
million).18 Consequently, the Government of Tamil Nadu undertook various efforts to address the
cataract problem.

The TNSBCS was established in April 1996 as an independent entity under the aegis of the National
Programme for Prevention and Control of Blindness. Its role was to give an impetus to the efforts of
the program by monitoring various chronic eye disorders and planning and executing schemes to
address them at the district level. Consequently, a World-Bank-assisted cataract blindness control
project was implemented in 1996 with a total investment of INR 0.64 billion,19 establishing district-
level blindness control societies across Tamil Nadu. Eye camps and district-level mobile ophthalmic
units were organized to detect chronic eye disorders. Alongside government-run district community
hospitals, the TNSBCS provided financial assistance to encourage voluntary organizations to perform
cataract surgeries.

The TNSBCS-led cataract treatments were successful; the state met a target of about 0.35 million
surgeries a year until 2001. However, the growing population, inadequate health infrastructure, and
sharp urban–rural divide motivated the TNSBCS to address the cataract problem by using emerging
technologies. As a result, in December 2020, the TNSBCS collaborated with the TNeGA to integrate AI
with healthcare. An immediate project conceived under this collaboration was developing AI for
cataract screening.

The TNeGA was a technology and innovation organization under the Information Technology wing of
the Government of Tamil Nadu, and its mission was to accelerate e-governance in the state. As a state
nodal agency, it had managed several initiatives such as developing low-cost, technology-enabled
scalable solutions in primary sectors such as education, healthcare, and agriculture. Its recently
established center of excellence aimed to promote solutions based on emerging technologies such as
AI, blockchain, and the Internet of Things (IoT) to address governance-related challenges. Their focus
had been the creation of a G2C service delivery system for barrier-free access to governance by the
commoner.

In 2019, the TNSBCS had performed about 0.2 million cataract surgeries against a target of 0.4 million.
Although the district tertiary hospitals had sufficient infrastructure, the prevalence of mature cataract
cases drastically increased among the elderly population residing in the rural pockets of the state.
With the mission of building cataract-backlog-free districts in Tamil Nadu, the TNSBCS approached the
TNeGA in December 2020 to jointly embark on an initiative to address the pressing need for screening
the rural aging population of the state using technology. In due course, the TNeGA devised a
compelling solution that was both affordable and easy to scale. Their partnership resulted in a
smartphone application that used AI to screen and identify people with operable cataract through
door-to-door screening by volunteers who had been given minimal technical training. The application
was named e-Paarvai.

18Population Census Data. (n.d.). Tamil Nadu population 2011–2023. Sex ratio & literacy rate 2023. Retrieved February 20,
2023, from [Link]
19 National Health Mission Tamil Nadu. (n.d.). National Programme for Prevention and Control of Blindness (NPCB).

Retrieved February 20, 2023, from [Link]

programme-for-prevention-and-control-of-blindness

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 4

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
KANMANI’S CATARACT DIALOGUE20

On a summer afternoon in 2021, Kanmani, a 63-year-old resident of Nagapattinam village in Tamil

Nadu, traveled to visit her close kin in Thoothukudi, a neighboring district. She was among the first
few to undergo cataract removal surgeries in her rural community, thanks to timely detection by the
e-Paarvai application.

To overcome the shortage of ophthalmologists, the TNeGA deployed e-Paarvai so that the TNSBCS
could detect cataract using a photograph taken with a simple smartphone (see Exhibit 4 for TNeGA’s
e-Paarvai Screening Application). Nagapattinam, Cuddalore, and Dharmapuri were among the top
districts where e-Paarvai was extensively deployed in 2021 to identify cataract and schedule the
subsequent surgical procedures (see Exhibit 5).

At Thoothukudi, Kanmani met her cousin, Govindarajan, and learned about his blurred vision problem.
She found that he was reluctant to get his eyes clinically evaluated. Among other reasons,
Govindarajan was concerned about the time that may be lost in traveling a long distance away from
his village to meet a doctor. Being the sole breadwinner of his family, he was apprehensive about
losing a day’s wages. Kanmani felt the need to take him through her journey with e-Paarvai.

“I too suffered the blurred vision problem since 2019, and later when the e-Paarvai volunteers visited
my village for eye screening, I was told that I was cataract affected and needed immediate surgery,”
said Kanmani. She went on to explain the simplicity and effectiveness of her overall experience.
“Volunteers from the TNSBCS came door to door to click pictures of our eyes, and there was no need
to go anywhere to get the eyes checked, and within a few minutes, I was astounded to hear that I had
cataract.”

Kanmani checked if he would be comfortable with a diagnosis obtained using a simple procedure: all
he had to do was look at a smartphone lens. He replied, “Maybe, but Kanmani, what exactly do you
mean by the word ‘cataract’ here? I have never heard about it.” Kanmani clarified: “Cataract is an eye
disease that causes clouding of the eye’s normally clear lens, and if it is not addressed, permanent
blindness could result.” Govindarajan was curious to know how a smartphone could detect such a
critical eye disorder.

The next day, Kanmani and Govindarajan met Madhavan, a local TNSBCS volunteer, to explore the e-
Paarvai process. After listening to Govindarajan’s questions about e-Paarvai, Madhavan saw an
opportunity to conduct an in-depth awareness session for the community residents in Govindarajan’s
area.

THE E-PAARVAI LEARNING SESSION

Around 200 residents of Thoothukudi gathered for the awareness session. During the meeting,
Madhavan explained that the TNSBCS aims to eradicate cataract by detecting it in the aged population
through the e-Paarvai application free of cost. Subsequently, he explained the harmful effects of not
addressing cataract and how using a smartphone with e-Paarvai could help detect cataract in their
eyes. He introduced his team of volunteers who visited homes, using e-Paarvai to perform eye

20 Fictional names have been used for the characters in the case.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 5

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
screening for residents aged 50 and above in the districts. “If cataract is detected, you might need
surgery at the local community hospital,” added Madhavan.

“We are happy that you do the screening at our doorstep free of cost, but how about our travel to the
hospital, which is far away from our houses, and the associated travel costs?” asked a listener. “The
TNSBCS covers the travel costs, and our volunteers will accompany you to the hospital to facilitate the
whole process until you return home after surgery. It is crucial to undergo eye screening and
subsequent surgery if cataract is detected, as the price of blindness far outweighs the loss of a few
days’ wages,” responded Madhavan.

“How long would it take from surgery to post-care?” asked another listener. Madhavan clarified: “The
screening-to-diagnosis phase would take around a week, followed by the surgery. After the surgery,
in less than 15 days, you will be free from cataract and can resume your daily activities.”

As the conversation moved forward, a resident, reluctant to bear any expenses for the checkup,
questioned the costs associated with the surgery and medications. The resident explained: “We earn
around INR 273 a day to make a living but to incur out-of-pocket health expenditure would be quite
challenging.” The anxiety of the crowd over the affordability of the process had to be addressed
immediately. Madhavan asked everyone to listen carefully, and exclaimed, “We have developed this
smartphone application to benefit senior citizens in the rural areas of the state, where the accessibility
and affordability of quality eye care screening were a challenge. Therefore, from screening at your
doorstep to surgery charges, all costs would be borne by the TNSBCS as part of the public healthcare
system.”

“You mean that my eyes would be diagnosed by a smartphone application and not a doctor directly?
What if I get misdiagnosed?” asked Govindarajan. Madhavan replied, “If anyone tests positive for
cataract through the smartphone application, a certified ophthalmologist at the local community
hospital will cross-verify the diagnosis through a manual eye checkup. This is a mandatory step before
the surgery.”

Madhavan recalled that although their first pilot failed, they managed to improve their smartphone
application’s accuracy. He said:

“We had our first field trial planned in Madurai, at Aravind Eye Hospital, in mid-
January 2021 with a simple model. A lot of data was gathered, and an application was
developed. We had to know why the first pilot had failed, and so we screened around
700 patients registered under the TNSBCS. We realized we had missed standardizing
the information. The screening was done hands-free using a smartphone, which led
to errors in the cataract diagnosis. It was later rectified, and after the correction, we
gained an overall accuracy level of 94%, which was a proud moment for our entire
team.”

Further, about 700 eye images were captured to create the dataset, which was trained using a
supervisory machine learning algorithm. A standard operating procedure was developed for
photographing the eye, and it was provided as a small manual inside the smartphone app. The training
dataset included an equal number of males and females aged 40 to 70 years. The pilot screening
covered 14 districts of the state. About 243 mature cataracts with very high accuracy (98%) and 380
immature cataracts with a marginally lower accuracy were detected. The team of certified
ophthalmologists appointed by the TNSBCS annotated the data collected in a district hospital in

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 6

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Madurai, Tamil Nadu, and an overall accuracy of 94% was achieved for the AI model (see Exhibit 6 for
the workflow of e-Paarvai).

Madhavan added, “The application was made available on the Google Play store platform and was
easily downloadable by the TNSBCS volunteers. It used computer vision and a state-of-the-art object
detection model. The images of the eyes were then clicked using a smartphone. After the volunteers
logged in, the patient details and images of both eyes were uploaded. The AI algorithm then processed
the images and confirmed whether the person had a mature/immature cataract or was cataract free.
Within a few minutes, the AI algorithm showed this categorization output in a dashboard that a local
community hospital could access” (see Exhibit 7).

A Web-based digital dashboard was built to validate the results. The district medical
officer/ophthalmologist at the government hospital conducted a comprehensive assessment of each
eye of the patient, approved the result, and then performed the surgery. The patient was called in for
surgery if the AI prediction was valid; otherwise, the data was stored automatically and used to train
the AI algorithm (see Exhibit 8 for more information on the application model).

For the cataract-positive patients, the local hospital was accountable for the surgery and updating the
dashboard (see Exhibits 9.1 to 9.4). The TNSBCS ensured a 30-day turnaround time from detection to
surgery.

Madhavan concluded by narrating the experience of a 100-year-old woman resident who suffered
from blurred vision problems, was successively diagnosed, and underwent cataract surgery facilitated
by the e-Paarvai team. He added that the e-Paarvai application detected a broken intraocular lens21
implant in her other eye during the follow-up checkups. The older lens that had been implanted in the
other eye during her cataract surgery done elsewhere long ago was found torn. This timely detection
by the e-Paarvai team ultimately saved her from possible blindness. This was an example of the e-
Paarvai solution’s reliability (see Exhibit 10).

KANMANI SHARES HER EXPERIENCE

Madhavan requested Kanmani to share her e-Paarvai journey with the audience in Thoothukudi.
Kanmani said, “I understand it is challenging to take leave from work but understand what will happen
if your vision worsens and one day you go blind. Unimaginable, right? Needing assistance and lifelong
dependency when you lose vision is far more difficult than not seeking help.”

She quoted a famous proverb: “Eyes are the window to the soul, meaning only our eyes convey our
emotions and thoughts,” and added: “You can barely manage to do anything with poor vision. We
should try and prevent the later consequences of misjudgments and poor decisions related to our
health. Fortunately, I have regained my clear eyesight through e-Paarvai, which has given me the
incentive to improve my life, and I must say it was a smooth experience.”

Further, she highlighted that if the whole community came together, blindness could be eradicated
from their midst, which would benefit society. She urged them to spread the word and reach a mutual
understanding among themselves to overcome the crisis that had been affecting them for decades.

21An intraocular lens implant is an artificial lens placed inside the eye that replaces the natural lens, which is surgically
removed, usually as a part of cataract surgery. Intraocular lenses are subject to wear and tear over time and can be
surgically replaced.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 7

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368

A few weeks later, Govindarajan tested positive for an operable cataract. Days after the surgery, he
visited Kanmani to convey his gratitude. He said:

“Thank you for educating me about the severity of the situation and introducing me
to the e-Paarvai team that day, Kanmani. Of late, I have been spreading the word
among my community members and encouraging them to get e-Paarvai screening
done. Indeed, it’s a great initiative by the Government of Tamil Nadu to save the
eyesight of millions.”

Kanmani and Govindarajan wondered about the social impact it would create if e-Paarvai were scaled
up across the entire state of Tamil Nadu and, further, throughout the whole country.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 8

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 1
NASSCOM AI Gamechangers Award bestowed on TNSBCS & TNeGA for e-Paarvai

Source: The Tamil Nadu e-Governance Agency (TNeGA) team.

Exhibit 2
WHO Definitions of Blindness and Visual Impairment
Blindness Presenting visual acuity <3/60 in the better eye with available
correction
Severe visual impairment (SVI) Presenting visual acuity < 6/60 – 3/60 in the better eye with
available correction
Moderate visual impairment (MVI) Presenting visual acuity < 6/18 – 6/60 in the better eye with
available correction
Moderate, severe visual impairment (MSVI) Presenting visual acuity <6/18–3/60 in the better eye with
available correction
Visual impairment (VI) Presenting visual acuity < 6/18 in the better eye with available
correction
Source: IAPB (The International Agency for the Prevention of Blindness). (n.d.). Definitions. Retrieved February 20, 2023,
from IAPB Vision Atlas website: [Link]

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 9

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 3
NPCBVI Data Platform with about 90% of Ophthalmologic Patients registered for Cataract

Source: National Programme for Control of Blindness & Visual Impairment (NPCBVI).
[Link]

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 10

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 4
TNEGA’s e-Paarvai Screening Application

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 11

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 5
Healthcare Workers using E-Paarvai App for Cataract Screening

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 12

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 6
Application System Workflow

Field worker uses an Android app to take images of a patient’s eye

If the patient is found is found to be cataract positive, then the patient details
are stored in the Dashboard

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 13

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368

The district administration will contact cataract-positive patients, cross-verify the diagnosis, and arrange for surgery

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

Exhibit 7
AI Predictions on the images captured

Prediction by AI Type

Positive Mature and Immature Cataract

Negative IOL and No Lens Distortion

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 14

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
Exhibit 8
About the Application Model
A multi-class classification model was used for classification, and the solution was deployed using Python
algorithms. The algorithms included a range of compound-scaled object detection models trained on the
datasets with bounding boxes and segmentation. The framework had a simple functionality for test time
augmentation, model ensembling, and hyperparameter evolution. Positive results were achieved by creating
sufficiently large and well-labeled datasets.

Exhibit 9
9.1. E-Paarvai Admin Dashboard

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 15

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368
9.2. e-Paarvai Centralized Dashboard showing all patient screening details

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 16

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368

9.3. e-Paarvai Dashboard: Data Validation by a doctor

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

9.4. e-Paarvai Dashboard: Patient Status Update

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 17

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.
 ISB368

Exhibit 10
100-year-old woman detected with broken Intraocular Lens Implant with her surgeon

Source: Officials of the Tamil Nadu e-Governance Agency (TNeGA) Team, Tamil Nadu, India.

An Artificial Intelligence (AI)-based Disruptive Innovation in Cataract Screening: The case of e-Paarvai | 18

This document is authorized for use only in Prof. Vijay Venkatraman's PGP-AI Strategy, 2025 (VV) at Indian Institute of Management - Bangalore from May 2025 to Aug 2025.