Clinical Trial Screening Scripts With ML Question Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Creating data-collection scripts for clinical trials is challenging due to the need to balance detail and simplicity, ensure accuracy, minimize false negatives and positives, and optimize participant engagement, which existing manual processes struggle to address efficiently.

Innovation Solution

Utilizing machine-learning models, particularly transformer models, to generate data-collection scripts that optimize participant engagement, reduce cognitive load, and enhance accuracy by incorporating evaluation and burden metrics, leveraging retrieval-augmentation generation systems for supplemental resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual processes are used to create data-collection scripts, then flexibility and customization are maintained, but development time is excessive and accuracy is compromised

Engineering Contradiction:

Improvescript development timeVSAvoidaccuracy of data-collection script

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

A machine-learning model acts as an intermediary between the clinical trial protocol and the data-collection script, automatically generating accurate and compliant scripts while reducing development time. The model is trained on historical protocols and scripts to learn best practices and regulatory requirements, serving as a bridge that maintains accuracy while dramatically improving productivity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The manual mechanical process of creating data-collection scripts is replaced with an automated machine-learning system. This substitution eliminates the time-consuming manual drafting and review process while maintaining or improving accuracy through the model's ability to consistently apply regulatory guidelines and best practices across all generated scripts.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If detailed questions are included to ensure accuracy, then false negatives and positives are minimized, but participant engagement and completion rates decrease

Engineering Contradiction:

Improveaccuracy in identifying eligible candidatesVSAvoidparticipant engagement

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The data-collection script is segmented into multiple sections or modules that can be dynamically assembled based on the specific clinical trial requirements and candidate responses. This segmentation allows the system to present only relevant questions to each candidate, reducing the perceived burden while maintaining comprehensive screening accuracy through strategic question selection and adaptive branching.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The data-collection script becomes dynamic rather than static, adapting its content and structure based on candidate responses, trial priorities, and eligibility criteria. The machine-learning model optimizes question sequencing and selection to maintain engagement while ensuring all necessary eligibility assessments are conducted, allowing the script to evolve based on real-time interactions.

Inventive Principle:

Principle #15Dynamics

3Reliability

If comprehensive screening questions are used to minimize false negatives, then all eligible candidates are identified, but the cognitive load and burden on participants increases

Engineering Contradiction:

Improveidentification of eligible candidatesVSAvoidcomplexity of data-collection script

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system applies partial action by selectively including only the most critical screening questions based on the specific trial's eligibility criteria and risk profile. The machine-learning model analyzes the protocol to determine which questions are essential for minimizing false negatives while excluding redundant or low-value questions, achieving comprehensive screening with reduced complexity through prioritized question selection.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20260018262A1Machine-learning techniques for generating data-collection scripts

Publication Date: 2026.01.15 TRIALSPARK INC D B A FORMATION BIO

AI summary

Disclosed embodiments may provide techniques for generating data-collection scripts. A computer-implemented method can include accessing input data. The computer-implemented method can also include processing the input data using a machine-learning model to generate a data-collection script. The data-collection script can include a set of questions for screening a plurality of candidates associated with the particular clinical trial. The computer-implemented method can also include generating evaluation metrics associated with the data-collection script. An evaluation metric can be associated with a particular question of the set of questions. The computer-implemented method can also include generating a modified data-collection script based on the evaluation metrics. The modified data-collection script can include a subset of the set of questions. A question of the subset can include an evaluation metric that exceeds an evaluation-threshold value. The computer-implemented method can also include transmitting the modified data-collection script to candidates associated with the particular clinical trial.