Rajashekhar C

RAJASHEKHAR C

Lead / Principal Product & Experience Designer

Design Leadership • UX Strategy • AI-Driven Product Experiences

Product and Experience Design professional with 15+ years of experience designing complex digital products, data-driven dashboards, enterprise platforms, and meaningful user experiences across healthcare and technology.

Intelligent Literature Screening System

AI-Assisted Safety & Compliance Workflow for Healthcare
Agentic AI

Designing an AI-assisted workflow to streamline regulatory literature screening, reduce manual review effort, and improve safety monitoring efficiency in life sciences.

Role: Lead Product & Experience Designer

Focus Areas: Enterprise UX · AI-Assisted Workflows · Healthcare Compliance · Workflow Optimization · Information Architecture

The Challenge

Pharmaceutical and medical device companies are required to perform continuous literature monitoring to identify adverse events, clinical outcomes, and product safety concerns. This process involves reviewing thousands of scientific articles and publications every year.

The existing workflow relied heavily on manual review processes, making it time-consuming, resource-intensive, and difficult to scale efficiently.

The goal of this initiative was to explore how AI-assisted workflows could streamline literature screening, improve operational efficiency, and support faster identification of relevant safety information.

The literature monitoring process required teams to:

  • Collect articles from multiple databases
  • Perform duplicate validations
  • Review clinical relevance manually
  • Verify target device references
  • Identify adverse events and performance issues
  • Prepare reports for compliance workflows

For organizations managing large product portfolios, this created:

  • High operational costs
  • Reviewer fatigue
  • Long onboarding cycles
  • Inconsistent review efficiency
  • Increased compliance pressure
Understanding the User
Workflow Overview

Research & Discovery

Understanding the Existing Workflow

The existing literature review process involved a combination of human reviewers and system-assisted checks.

Key screening stages included:

  • Language validation
  • Duplicate detection
  • Primary clinical data verification
  • Target device identification
  • Clinical outcome validation

The workflow revealed multiple repetitive decision-making steps that contributed to cognitive overload and review inefficiencies.

UX Challenges

Key UX Challenges Identified

Information Overload

Reviewers processed hundreds of articles per review cycle.

Repetitive Manual Validation

Duplicate checks and inclusion/exclusion verification required repeated effort.

Cognitive Fatigue

Constant context switching reduced screening efficiency.

Trust in Automation

AI recommendations needed to remain explainable and reviewable.

Workflow Scalability

The system had to support large product portfolios and evolving regulatory requirements.

Solution

Strategy Section

Design Strategy

The solution approach focused on creating a human-centered AI-assisted review workflow rather than fully replacing human decision-making.

The platform aimed to:

  • Reduce repetitive screening effort
  • Improve review prioritization
  • Surface critical findings faster
  • Support explainable AI recommendations
  • Maintain regulatory review confidence
Solution

UX & Workflow Thinking

The design process involved:

  • Workflow mapping
  • Screening logic analysis
  • Decision-point evaluation
  • Inclusion/exclusion criteria breakdown
  • Human-AI interaction planning

The screening workflow became a central part of the experience design process.

Solution

Proposed Solution

An AI-assisted literature monitoring platform designed to:

  • Automatically collect and organize literature
  • Detect duplicates intelligently
  • Identify inclusion and exclusion criteria
  • Flag adverse events and performance issues
  • Prioritize relevant articles for reviewers
  • Generate summarized annotations for faster analysis

The platform combined AI automation with human validation to improve both speed and reliability.

Workflow Overview

Key Features Section

Key Product Features

AI-Based Screening Assistance

Automated article classification using predefined inclusion and exclusion logic.

Intelligent Duplicate Detection

Reduced repetitive reviews by identifying similar publications.

Clinical Relevance Identification

Highlighted articles containing target device references and adverse event indicators.

Reviewer Dashboard

Centralized review workflow with prioritization and article tracking.

Annotation & Summarization

Enabled faster interpretation of lengthy scientific publications.

Compliance-Oriented Workflow

Supported structured review processes aligned with regulatory expectations.

Impact

Expected Impact

The proposed workflow aimed to significantly reduce manual effort associated with literature screening.

Potential benefits included:

  • Reduced operational costs
  • Faster article review cycles
  • Reduced training dependency
  • Improved review consistency
  • Faster escalation of safety-related findings

In large-scale review programs, organizations screened tens of thousands of articles annually, creating strong opportunities for workflow optimization and efficiency improvements.

Impact
Key Learnings

Key Learnings

Designing for Human + AI Collaboration

A major focus of the experience design was ensuring that AI-supported recommendations remained transparent and understandable for reviewers.

Rather than replacing human expertise, the experience was designed to:

  • assist decision-making
  • reduce repetitive effort
  • improve confidence in screening workflows
  • maintain reviewer control over final outcomes

Special attention was given to:

  • explainable recommendations
  • workflow clarity
  • confidence indicators
  • reduced cognitive load

Key Learnings

This project highlighted the importance of balancing AI automation with human expertise in regulated healthcare environments. The experience reinforced how enterprise UX goes beyond interface design — requiring deep understanding of workflows, operational constraints, compliance considerations, and decision-support systems.

Impact & Outcomes

This project helped me evolve from designing interfaces to designing systems that integrate AI, user behavior, and business workflows.