endorsedOngoing

Predictive ML Platform for Cancer Pathway Capacity Planning

Services Covered:

Custom Software Development • AI and Machine Learning • Big Data Analytics and Business Intelligence

Industry

Business Consulting & Services

Duration

80 months

Budget

10K - 49K

Start Date

1 December 2019

Team Size

1-5 Employees

Engagement Model

Nearshore

Project Capability Score (PCS)

71/100

Good

Project Capability Score (PCS) estimates how capable Neoteric is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Business Intelligence Group

London, United Kingdom

Project Summary

Neoteric validated an initial hypothesis and built a predictive machine-learning platform from the ground up using Python and Open Library, supporting continuous data calculation and presentation via Qlik and SQL Server. The resulting model achieved strong initial accuracy, and the vendor continues to support and refine the product in an ongoing engagement.

Key Challenges

Reduce delays for patients on cancer pathways by enabling capacity planning
Analyze historical data from multiple organizations to build accurate predictive models
Develop an MVP that can be deployed across several organizations in the consortium

Project Deliverables

Custom machine learning algorithm/model
End-to-end software platform built from scratch
Data visualization and reporting integrations (Qlik, SQL Server)
Pilot deployment to a single organization

Project Solution

Neoteric first ran tests to validate the hypothesis and built an initial algorithm. After scoping sessions (including an on-site two-day workshop in Poland), they developed the product from scratch using Python and Open Library, implemented statistical models (including linear regression), and integrated data presentation via Qlik and SQL Server. They provided a core team (developers, data scientist, project manager) and ongoing support for refinement and deployment.

Project Outcome

Achieved just over 80% accuracy with approximately a 19% error rate on the initial model
Deployed as a pilot in one organization with plans to roll out to the remaining organizations
Produced model predictions that were close to real-life data and continue to be refined with additional years of data

Platforms

AI/ML Platform
SaaS

Tech Stack

Python
SQL

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Neoteric created a robust model with an incredibly high accuracy rate and a low error rate. The entire team continues to be supportive, providing excellent customer service and going the extra mile to deliver great results. They are a communicative, accommodating, and knowledgeable partner.”
Femi Odewale
CEO
Date: 04/03/2020