Towards increased realism of a computer simulation of human childbirth.

Sadulaev, Said-Magomed (2019) Towards increased realism of a computer simulation of human childbirth. Doctoral thesis, University of East Anglia.

[img]
Preview
PDF
Download (19MB) | Preview

Abstract

A virtual childbirth simulator is normally associated with a computer analogue of mechanical mannequins used for training purposes in obstetrics. Such a simulator would allow acquiring a deeper understanding of labour and provide the necessary expertise for students in obstetrics. A patient-specific childbirth simulator, in turn, would be capable of predicting difficult birth scenarios in advance based on ultrasound or magnetic resonance imaging scans of the maternal pelvis and fetus. This would give midwives and obstetricians time to prepare for predicted worst­case scenarios and potentially reduce morbidity and mortality of both babies and their mothers. The existing virtual childbirth simulator successfully simulates physiologic labour. This thesis is concerned with taking the software one step closer towards being a patient-specific virtual childbirth simulator and to simulate difficult birth scenarios.

The core content of this thesis is concerned with the development of computational feta I neck models. A number of neck models were developed and tested in the simulator. The methods used to simulate the feta I neck are the following: ball and socket joint for intervertebral discs, spring-and-damper systems for ligaments and six-degrees­of-freedom bushing element to simulate a coupled behaviour of the discs, ligaments and neck muscles. The latest one-pivot neck model is using a six-degrees-of-freedom bushing element to simulate the behaviour of the feta I head. The developed neck model, together with the approximated complete mechanical properties of the feta I spine, facilitated running the experiments with a higher variety of biomechanical parameters such as the neck's length, strength and a full range of feta I bi parietal diameters. The experiments are reported in this thesis.

An additional simulation software, using haptic devices, was developed specifically for validation of the developed computational neck models. The software allows manipulating of a virtual feta I head on the screen, using two haptic devices. It is used to validate the resistance of the feta I head, during flexion, extension and rotation. It was clinically tested by midwives and obstetricians at the hospital. The results showed that the software is capable of providing biomechanical properties of a newborn's head motion, with the help from the clinicians.

Experiments were conducted to validate the accuracy of the Total Lagrangian Explicit Dynamics (TLED) contact method, used in the software. The validation setup consisted of a finite element cube and a rigid body plate, pushing vertically down on the cube with a gravitational force of 9.81N. Similarly, the experiment was repeated for a rigid body sphere pushing on the top of the cube. The results showed that TLED is less sensitive to the number of tetrahedral elements as compared to the Abaqus Explicit contact method.
Another set of experiments were conducted for resolving a direct occipito-posterior position (OP) of the fetal head, which is considered to be a difficult birth scenario. In OP, midwives advise their patients to tilt their pelvises anteriorly to help with labour. This method was experimentally tested in the childbirth simulator and the results showed that tilting a pelvis anteriorly could potentially ease the dilation during the first stage of labour. However, no significant difference was observed as compared to the non-tilted pelvis during the second stage of labour.

Experiments were run to observe shoulder dystocia in the childbirth simulator. Initially it was not possible to observe shoulder dystocia due to the rigid fetal trunk coming into contact with the maternal sacrum. A number of adjustments were made in order to allow the rigid feta I trunk to follow the feta I head, such as increasing the feta I shoulders to the average width, cutting the trunk in half and disabling the spring keeping the trunk vertical. Shoulder dystocia was observed in the simulator even in the absence of the complete birth canal, articulated fetal shoulders and flexible fetaI chest.

Finally, a new maternal pelvis was introduced in the software with a mobile sacrum. The sacrum was attached to the rest of the pelvis using the six degrees-of-freedom bushing element. The effects of the sacrum mobility on a childbirth were studied. The results show that a mobile sacrum contributes toward the full internal rotation of the fetaI head during a childbirth, whereas the absence of the mobility leads to the arrest of the head in the anteroposterior diameter of the pelvis.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Science > School of Computing Sciences
Depositing User: Chris White
Date Deposited: 07 Apr 2021 13:44
Last Modified: 07 Apr 2021 13:44
URI: https://ueaeprints.uea.ac.uk/id/eprint/79638
DOI:

Actions (login required)

View Item View Item