top of page
Mac Studio.png

Desktop application for pathologists in analyzing and reporting blood smears.

Roles & Responsibilities

0 -> 1 Product Design - Feature Scoping, Research, Interaction Design, Visual Design, Prototyping

The Team

2 Product Managers

3 Developers

1 Product Designer (myself)

​

Duration

January 2022 - May 2022

Overview

The rising testing volume, increasing patient complexity in clinical labs, and inter-user variability in test results due to manual methods and equipment discrepancies, highlight a multifaceted problem in hematology diagnostics. To address these challenges, my team and I aimed to enhance efficiency, accuracy, and accessibility in hematology analysis, benefiting pathologists.

891840.png

Rising testing volume

With the advancement of diagnostic and therapeutic techniques in hematology, more people are being diagnosed with cancer and other blood-related diseases.

3974825.png

Increasing patient complexity

The world population is also expanding and aging. This means people are living longer and developing complex conditions with comorbidities.

pathology-icon-490x512-dpmvwhor.png

Inter-user variability

Traditional hematology laboratories rely on experienced staff to make consistent clinical decisions using manual instruments, leading to the variability of test results.

Problem Framing

How might we streamline hematology diagnostics and enhance the workflow for pathologists?

The Solution

Designing a stand-alone application for pathologists to manage the patient workflow digitally,

  • Improving turnaround time

  • Optimize usability of the application

  • Reducing human error 

Mask groupk.png

A slide put under the "scanner" and then calibrated to view it on the desktop

Mask group.png

The digitized images enables the pathologist to virtually “see” the slide as if it were under a microscope

User Interviews

I conducted user interviews with stakeholders invloved to better understand their pain points, behaviour and motivation while working with the conventional method of heamotology

My objectives included understanding workflow challenges, timelines, patient list organization, pathology ecosystem familiarity, common practices, and navigation preferences.

" 

" 

Insights & Analysis

Users often feel overwhelmed by the high volume of cases handled daily with microscopes, leading to the need for revisiting cases without access to previous analyses.

Users possess knowledge of existing tools but lack a consolidated application that fulfills all their requirements.

Users express frustration due to the intricacies of using microscopes and challenges related to calibration for slide examination.

Users are enthusiastic about adopting innovative technology to enhance public health and improve patient outcomes, particularly considering the substantial variability in slide content.

Users are motivated by a desire to leverage advanced tools available in the market to enhance patient outcomes and healthcare quality.

User Persona

A MEDICAL PROFESSIONAL’ PERSONA

I created a fictional persona named Sheetal who fits the demographic of my representative users. I then proceeded to polish this persona by identifying her desires, needs, and frustations.

HMW Statements

Every problem is an opportunity for design. By framing my challenge as 'How Might We' questions, I was able to set myself up for an innovative solution.

  • How might we instill confidence in users regarding the accuracy and reliability of the digitization and analysis process?

  • How might we design an onboarding process that efficiently educates users?

  • How might we establish a robust approach to identify and present potential errors?

  • How might we provide users with greater visibility into the AI's operations, making it easier for them to comprehend its actions and decisions?

Information Architecture

Group 17.png

User Flow

Group 16.png

Testing + Improvements

Based on various feedback from 8 other peers, I continually iterated my design over the span of 4 weeks- with 3 major improvements:

Switching the design to Tab View

first it.png

First Iteration

  • As the Calibration, scan & analysis process takes a considerable amount of time, providing the user with a tab view can enable working on multiple cases at a time thus, increase in productivity.

  • Able to view the details of the case/slide (test type, stain type, etc) in a different tab, gives the user option to go through it clearly and frequently.

Other Improvements

dynamic notif.png

Dynamic Notifications

Dynamic Notification Bar

  • A dedicated space to show errors and notifications that keeps updating accordingly as the user performing the tasks

  • Approaching robust design is required for notification/errors in such a way it grabs user attention and can make the changes quickly and in mitigating the potential errors or any wrong attempts.

  • Also able to track the time while scanning and analysing.

Automated AI Process

  • The AI process (typically 2-3 minutes depending on the size of blood smear) has to be communicated clearly and effectively to the pathologist on the spot so that user can take action if necessary.

Final Screens

Group 19.png
Home Screen

Future Roadmap

  • Increase the capability of application with other tests, ex: Histopathology (examination of tissue).

  • Integrate with the web based workflow management tool for any further referrals to doctors or to examine.

  • Part of its normal workflow generates an abundance of free training data that can be used to train an automated Artificial Intelligence (AI) system. Understand, Refine and Establish more streamlined flow of AI process.

Learnings & Outcome

  • This internship taught me a lot about embracing ambiguity, asking the right questions, how to ship products at a scale, and finding impact.

  • Be cautious, but never assume- As it is my goal to design with as much empathy as possible, this project made me learn to be extremely sensitive with topics that deal with healthcare or AI.

  • The importance of user feedback- Being the only designer is a double-edged sword. This makes collecting user feedback extremely critical in the process.

  • Iterate as much as you can- In the beginning stages, I’ve explored so many different options to try finding the right solution for my users- I’ve ended up “restarting” the project over 3 times.

Other Projects

Group 12.png

Optimizing digital pathology workflows by enhancing case management and streamlining the slide viewing platform.

Onward Health 2022-23, UX lead

Group 14d.png

Mobile viewing experience for oncologists who are reviewing the interpretation done by radiologists.

Qure.ai / Design Challenge 2022-23

bottom of page