Clinical Trials

Does clinical research create more and more data, are we making the most of it, or are we in danger of being overwhelmed by the enormous amount of data?

Using data as a resource and exploiting advanced analytical and cognitive approaches in decision-making therefore becomes the focus for future financial sustainability in the sector.

Ensuring reliable times and costs.

Also using predictive Risk Based Monitoring techniques, with anomaly detection and predictive modeling.

 USE CASES

  • Site Selection (demo available)
  • Process Mining (demo available)
  • Drop-Out
  • RBM (demo available)

Site Selection

Powered by SAS Viya, a cloud-enabled memory analysis engine that provides fast, accurate and reliable analytical insights.
ML algorithms to identify sites that are most likely to perform better or worse during the clinical study.  
Faster processing of large amounts of data and more complex analysis for faster study management.

Monitoring sites based on data.

It identifies centres that perform best in meeting protocol times.

Improves collaboration in the study, allowing better visibility on critical cross-site data, comparing trends and patterns, qualitative and quantitative kpis.
It reduces manual processes, reducing the burden on study teams and potentially lowering study costs.

Process Mining

Identification of bottlenecks and inefficiencies. Control and optimization of the different phases of a process.
Network analysis, aim to highlight how the study is carried out in the various stages and then detect anomalies.
Generate graph where line thickness is given by the standard deviation of atraversamento times. Respected times, low lines.
Determine the person towards Eot = Enf od treatments or Eos = End of Study.
Reports or analysis support allow you to explore the data
Also thanks to decision trees. Decision Tree.
Algortimo updates with changing data. 

Supervised ML algorithm, not always for predictive purposes even exploratory type.

Drop-Out

Supporting analysis to predict which and how many patients are most likely to leave or not to follow study protocols and timelines.

Historical series analysis, case evaluation to determine the probability of completion of the study in the terms and times assumed.
It allows clinical trials to be managed more efficiently, reduces costs and minimizes unanswered milestones.
It uses supervised ML algorithms.

RBM

Risk Based Monitoring methodology.
Map performance, process and risk indicators. Generation of synthetic and predictive indicators. Possible integration of BSC methodology.
ata visualization attractive and intuitive, navigation in hierarchy and not.
Use Risk Detection methodology, exploit the power of algorithms to predict risk and future trends for basic or synthetic indicators.
Quickly identify patterns, trends and relationships in structured and unstructured data.  
Turn data and Kpis into real-time usable information.
RBM is now a methodology also requested by the FDA Food & Drug Administration.

Some Benefits

Reduces manual processes, reducing the burden on study groups and potentially lowering study costs

Improve collaboration, enabling better visibility, transparency and sharing of critical and non-critical data

Turn data into real-time actionable information

Interact with data through a centralized, easy-to-use, cloud-based platform

Manage clinical trials more efficiently, reduce costs and minimize key goals lost

Create an historical E-DWH and allows cross-trial security queries to run

View clinical data before sending using integrated security and integrated analysis and detail or synthesis

Main needs addressed to:

  • 80% of ClinOps report that they regularly miss milestones
  • As a result, every ClinOps leader reports that they have initiatives in place to improve milestone achievement
  • ClinOps leader agree that the following four initiatives are critical for improving success in focus areas and reducing the incidence of missed milestones:
    • Study quality metrics
    • Continuous quality
    • Risk-based monitoring
    • Centralized monitoring
  • To reduce the likelihood of missed milestones, ClinOps leaders focus on three critical processes:
    • Enrollment
    • Site productivity
    • Subject compliance
  • Fewer than 10% of respondents report having access to the digital tools that enable automation in trial oversight
  • In addition, 60% of ClinOps leaders report they use manually compiled spreadsheets for their operations
  • Reflecting the lack of appropriate tools for managing their clinical trials, these leaders described a number of limitations as the key barriers to improving the management of clinical trials :
    • Inability to investigate study issues in real-time
    • Lack of visibility into data and actions being taken to address issues
    • Data across too many disparate data sources makes it tough to identify issues
  • The majority agree that automation—in lieu of manual processes—for data collection, analysis, and study management is critical for improving milestone achievement
  • Most respondents identify real-time data aggregation across study systems  as the most important attribute of any solution

contact us to learn more