Data mining approaches to predicting ovulation and fertility problems

PhD project (3/4 yr research project leading to independent research at the doctorate level)

Dr Tom Gaunt, Prof Debbie Lawlor, Prof Toby Knowles, Dr Louise Millard


Return to list

Rationale

Many couples have difficulty conceiving, both female and male fertility problems are surprisingly common. A key to successful conception is the ability to track the menstrual cycle and the timing of ovulation. Intra-vaginal ultrasound has been used for this purpose, but a new equally accurate, but more convenient, method has been identified as an aid to the diagnoses of other underlying fertility problems. An accurate temperature sensor is worn internally overnight and records temperature every 5 minutes. The data are downloaded via a smartphone app to a central server for data processing and anonymous storage. The individual measurements are then used to predict and also to confirm ovulation for the user. Different patterns of change in the data can also be associated with a range of fertility problems/disease. These cycle monitoring products are commercially available and we have access to a large and growing database of anonymised overnight and day-summary data.

Aims & objectives

1. To initially identify different profiles of change within overnight/monthly temperature cycles.
2. To investigate associations between profiles of temperature change and specific fertility problems/disease.
3. To improve prediction and verification of the timing of ovulation from overnight/monthly temperature data.

Methods

The Studentship provides a valuable opportunity to gain skills and training in working with ‘big data’ at an applied level. The successful candidate will develop skills in epidemiology and the rapidly developing field of data science, specifically learning state-of-the-art statistical approaches to investigating functional data, data mining and epidemiology, as well as techniques to process large datasets and present / visualise findings in accessible formats. The student will also attend relevant internal and external training courses.

References


Created on Nov. 29, 2016, 1:21 p.m.

Edit this page