Background: Currently, there is no accurate method to objectively assess transtibial prosthetic socket fit. This study aims to evaluate the use of infrared thermography (IRT) to analyse temperature changes in response to pressure points, and their relation to pain and comfort assessments. This study also explores the reliability and agreement parameters of IRT.
Methods: A within-subject study was conducted on seven participants wearing a vacuum or pin suspension socket. Each participant was examined with two sockets: their prescribed socket (unmodified) and their prescribed socket modified with added pressure pads inside the prosthetic socket at the fibular head and distal tibia end. Thermal images were obtained after 10 minutes of walking with the unmodified socket (TW1), after a subsequent 15-minute resting period (TR1), after 10 minutes of walking with the modified socket (TW2), and after a second resting period (TR2). Difference maps showing temperature changes relative to the baseline (TR1) were created. Subjective assessments included localised pain and Socket Comfort Scores. Regions of interest (ROI) for temperature analysis were selected at the fibular head, distal tibia end, popliteal fossa, and gastrocnemius belly. Reliability and agreement were assessed using the Intraclass Correlation Coefficient (ICC), standard error of measurement (SEM), SEM%, smallest detectable change (SDC), SDC%, and Bland-Altman plots.
Results: Visual inspection and ROI temperature values revealed hotspots in 86% (6/7) of participants at the fibular head post-walking with the modified socket, contrasting with the unmodified socket in the majority of cases. Pain corresponded with the presence of hotspots at this location. Hotspots at the distal tibia end post-walking with the modified socket were observed in 43% (3/7) of participants, but without apparent difference from the unmodified socket. Only two-thirds of these coincided with reported pain. Reliability of IRT was very good to excellent across the residual limb and ROIs (ICC ≥ 0.85), except for fair reliability at the fibular head ROI (ICC 0.55). For TR1 and TR2 test-retest agreement, SEM ranged from 0.3 °C (1.0%) to 0.7 °C (2.1%), with SDC from 0.9 °C (2.8%) to 2.0 °C (5.9%). On the difference maps of unmodified and modified sockets, SEM ranged from 0.6 to 0.8 °C, and SDC from 1.7 to 2.1 °C.
Conclusion: Based on the observed thermal response to pressure in the majority of cases, IRT demonstrated potential in identifying peak pressure points and pain through increased temperature measurements at the fibular head, but had limited capability in detecting these at the distal tibia end. This limitation is likely due to differences in circulation, socket suspension effects, and uncertainty about the exerted pressure. Future research should simplify the protocol to enhance clinical applicability, enabling IRT to complement subjective socket fit assessments and improve prosthetic socket designs and patient outcomes.

Background Over the past few years, virtual reality-based visual field assessments have been increasingly investigated. However, due to the limited field of view (FOV) that a virtual reality (VR) headset can offer, research to identify the visual field's outer limits has not been possible as the FOV falls short of the human visual field limits. This limitation has hindered the ability to accurately identify the outer limits of the visual field using VR technology. The Peili Research app, a VR-based kinetic perimeter, addresses this issue by employing off-centred focus points to expand the effective angle that can be tested.
Research Objective This research explores the feasibility of employing the Peili Research app, alongside the Pico Neo 2 Eye VR headset, as a bedside tool for visual field assessment.
Methods A series of analyses were conducted to evaluate the Peili Research app’s performance in visual field testing. The Spearman correlation matrix was used to assess test-retest reliability, and a comparison with Goldmann perimetry was made. The app’s design was evaluated by calculating the increase in effective testable angle through the off-centred focus point. In addition, the impact of observer differences due to human reaction time and the impact of stimulus colour contrast on the results was investigated. Data processing was conducted using a custom-developed MATLAB script to analyse the Peili data and determine the visual field angles in isopters.
Results Visual representations of the Peili Research app suggested a certain extent of test-retest reliability. However, Spearman correlations varied between poor and moderate. The application extends the testable visual field angle by 23 degrees on average. The study revealed that human reaction time in recording visual stimuli on the tablet led to an average underestimation of visual field angles by 5.38 degrees (p<0.001). Furthermore, colour contrast significantly affected detection angles, with green stimuli being detected at larger angles compared to blue and yellow stimuli (p<0.001). Clinical testing on postoperative volunteers highlighted practical challenges.
Conclusion The Peili Research app shows promise as a VR-based tool for visual field assessment, capable of expanding the testable visual field angle. However, variability in results, mainly due to the location of the focus points, reaction time, colour contrast, and the instability of the VR headset, highlights the need for further design refinements. Addressing these issues could enhance the app’s clinical applicability, making it a valuable tool for bedside visual field testing and improving patient outcomes.

Wheelchair basketball has become increasingly popular, leading to a rise in professionalism. While performance measures exist, they lack objective metrics directly related to an athlete’s individual load that can be measured during games. An objective measure related to the athlete’s load can provide information about fatigue and the total load of training or matches. A recent study presented a theoretical framework for calculating power during games. This study aims to examine the utilization of power metrics derived from Inertial Measurement Units (IMUs) in wheelchair basketball using the theoretical framework, focusing on power produced during straight-line sprinting in matches. This will be done by answering the main question of this paper: How can
individualized power metrics for performance monitoring be derived during wheelchair basketball match play using IMUs? Eight female participants from the Dutch national wheelchair basketball team were assessed in twelve international practice games using
IMUs on their wheelchairs. Power profiles were created based on sprint power, offering insight into sprint powers and their distribution. Work done, determined from power output and push duration, provided insights into athlete fatigue during games.
Power profiles can be used to monitor long-term performance, either between games or between seasons. Regression analysis showed a significant positive effect of classification scores on single push power output, with an R-squared value of 0.75. This study
proposes areas for future research, including integrating trunk motion analysis and exploring the effects of different player positions on power profiles. By enhancing the understanding of player performance, these findings contribute to the professionalization of wheelchair basketball, aiming to optimize performance and reduce injury risks.

The ability to simulate the impact on performance of military body-borne loads enables effective analysis and optimisation of load and equipment configurations for military personnel performance. Additionally, these simulations can reduce research & development costs of new equipment by analysing its impact in an early stage of development. While research has been done into the effects of body borne loads on kinematics, ground reaction forces (GRF) and metabolic cost of transport, the accuracy and reliability of simulations are not clear yet.
This study set out to predict loaded gait kinematics and GRFs and estimate metabolic cost of transport for gait at 1.5 m/s, carrying different types of military relevant body-borne loads, to evaluate the applied methods for implementation in a future load configuration optimisation tool.
The kinematic/GRF prediction was performed by forward dynamics simulations in SCONE/hyfydy, using a planar musculoskeletal model and a 2 gait-state controller, simulating 15 solutions for each load condition. While only 60% of experimentally measured differences between load conditions were correctly predicted, the expected differences from literature were all correctly predicted. It was assessed that improving the number of gait states and the number of optimisations per load condition is expected to improve these results.
The metabolic cost of transport (mCoT) estimation was performed by the Computed Muscle Control algorithm of OpenSim, using experimental kinematics and ground reaction forces. However, small compounding errors in experimental data and data processing prevented accurate mCoT estimations. Although the applied kinematic/GRF and mCoT simulation methods could not be validated yet, based on these results, the study as a whole does show promise for the continued development of these models and their future implementation for loaded gait performance optimisation.

In today’s medicine, 3D planning and PSI are not commonly used in revision TKA surgery and little is known about the accuracy and implementation of these techniques in revision TKA. Therefore, the question arises, what are the potential benefits of 3D planning and PSI in revision surgery? The goal of this thesis is to explore the potential benefits of 3D planning and PSI in revision surgery. To achieve this, a retrospective study and a prospective study are performed, and a first design of a 3D printed PSI is created. For both retrospective and prospective study, 3D plannings are made of patients undergoing revision TKA, comparing the pre-operative planning with the post-operative results based on size prediction, augmentation prediction and component placement analysis. The results of the studies conducted in this research align closely with each other, providing similar results for size and augmentation prediction and component placement analysis. All in all, these studies indicated consistency and reliability of the 3D prediction in different scenarios, affirming the potential of 3D planning in revision TKA surgery. Additionally, the first design for a PSI, a 3D-printed guide, is constructed. This design incorporates several important landmarks for component placement during revision TKA, increasing the outcome of these surgeries, especially if landmarks are missing or unidentifiable during surgery. Together 3D planning and PSI have a bright future in revision surgery, however, future research needs to be conducted to implement 3D planning and PSI in modern-day healthcare.

Introduction: Deep brain stimulation (DBS) is an important therapeutic option for various neurological diseases. For certain indications the optimal target cannot be identified on structural magnetic resonance imaging (MRI), but can be visualized with tractography. A recent improvement for clinical practice is the probabilistic tractography algorithm, providing superior results to the deterministic counterpart. However, it is suggested that clinical users are unfamiliar with the complex technology of this new approach, which might lead to missed potential of tractography in DBS-care. The overall aims of this thesis were threefold: to create an overview of the national landscape on the application of tractography, to narrow the gap between medicine and advanced technology in the application of tractography in DBS surgery planning, and to report potentialities and pitfalls of the implementation of tractography in a general hospital.

Methods: In the first part, a survey was conducted among Dutch DBS clinicians on the deployment of tractography. The survey consisted of 25 questions about the provision of DBS care, the use of tractography, the expert’s opinion and the respondent’s demographic characteristics. A comprehensive literature study is conducted in the second part. In the third part, an economic evaluation is performed by analyzing the costs and benefits of tractography. Guidelines are suggested based on literature and expert experience.

Results: Tractography is considered valuable for essential tremor (p<0.001) and not valuable for epilepsy (p=0.002) and chronic cluster headache (p=0.016). The majority uses deterministic approaches like DTI. Probabilistic users consider tractography more valuable than deterministic users (p=0.036). There is a heterogeneity in used image acquisition parameters and a lack of knowledge exist on technical background of tractography. The key elements of the technical principles of tractography are summarized, including diffusion weighted MRI, the concept of estimating the diffusion tensors, image acquisition parameters and tract reconstruction with the deterministic and probabilistic approach and quantitative measures.

Conclusion: Tractography is used by DBS clinicians with limited knowledge on the technical aspects. Generally the inferior deterministic approach is used without an (inter)nationally standardized protocol for image acquisition and tract reconstruction. A comprehensive explanation of the technical concepts and suggested practical guidelines should contribute to better application of this promising technology in DBS-care.

Cardiothoracic surgery is a common treatment for cardiovascular diseases. Patients are admitted to the ICU after cardiothoracic surgery for continuous monitoring to prevent or treat postoperative complications as much and as soon as possible. One of these complications is the development of circulatory shock. It is likely caused by one or a combination of several factors, leading to increased morbidity and mortality in the ICU. The main purpose of the circulation is to transport O2 and nutrients to the tissues and remove waste products of the tissues via the tissue’s microcirculation. Under normal conditions, O2 supply exceeds O2 demand. However, during circulatory shock, the circulation cannot meet the perfusion demands of the organs, leading to organ dysfunction and organ failure. Resuscitation procedures for patients with circulatory shock focus on normalizing macrocirculatory parameters, such as CO and SvO2, by administering fluids and vasopressors to support tissue perfusion. Improvement in macrocirculatory parameters is expected to be paralleled by improvement in microcirculatory perfusion and tissue oxygenation (i.e., hemodynamic coherence), but it appears that these do not always improve simultaneously. Loss of this coherence has been associated with adverse outcomes. 

The microcirculation can be imaged sublingually with an HVM. Studies in patients with septic shock have shown that hemodynamic coherence is often lacking. Therefore, it could be valuable to monitor the microcirculation of cardiothoracic surgery patients. 

This thesis aimed to investigate the postoperative time course of microcirculatory parameters in patients admitted to the ICU after cardiothoracic surgery with and without circulatory shock,  the relationship between macro- and microcirculation, and the usage of leukocyte detection in understanding patient’s systemic inflammation.  

Chapter 2 provides general background information on cardiothoracic surgery, CPB, the physiology of the microcirculation, the latest generation of HVM, pathophysiological changes in the microcirculation after cardiothoracic surgery, leukocyte-endothelium interactions, the macrocirculation, and hemodynamic coherence. A retrospective study of cardiothoracic surgical patients with shock is described in Chapter 3 of this thesis. The results showed that the microcirculation might adapt to compensate for the circulatory shock state by decreasing RBCv and increasing FCD, TVD, and cHct compared with normal values of healthy volunteers while maintaining tRBCp. Chapter 4 describes a prospective study comparing cardiothoracic surgical patients with shock from Chapter 3 with cardiothoracic surgical patients without shock. The comparison between these two groups showed that both groups exhibited different behavior of the microcirculation. However, the underlying mechanism is not understood and requires further research. Chapter 5 contains an explanatory review of the use of STDs for leukocyte detection. Chapter 6 provides a general discussion and reviews the future prospects of microcirculation measurements as a tool in the management of critically ill patients. Our findings should be examined in more extensive clinical trials to determine whether microcirculatory changes contribute to the development of shock.

Severe muscle weakness is a symptom appearing in certain neuromuscular diseases (NMDs), such as Duchenne Muscular Dystrophy (DMD), affecting people's daily lives by reducing functionality, decreasing independence, and reducing the ability to perform essential daily activities. This patient group might benefit from using active-assistive devices by having the potential to provide precise support torque counterbalancing the passive forces acting on the arm, the movement intention of the user, and external forces exerted by lifted objects. However, the determination of support to counteract the weight of lifted objects is an ongoing challenge. This research aims to improve the understanding of external forces by using data classification algorithms to distinguish between different lifted weights in a human experiment. Fourteen healthy individuals participated in this experiment, lifting weights ranging from 0 - 1000 grams while an active-assistive device compensated for the passive torques acting on the arm. Data was collected using various sensors: a force sensor, an Inertial Measurement Unit (IMU), a joint encoder, and surface Electromyography (sEMG) electrodes. Subsequently, this data was processed and fed into a K Nearest Neighbour (KNN) classifier and a Support Vector Machine (SVM) classifier to determine the lifted weights during human elbow flexion and extension. The classifier showing the highest performance achieved an accuracy of 39.70% on the test dataset, indicating several misclassifications. However, a recall percentage of 76.95% for the 1000-gram class within the multi-class classification demonstrates the capability to distinguish larger weights. While demonstrating potential in weight discrimination, especially for larger weights, improvements in the compensation strategy, arm support alignment, and experimental design are crucial. Future research on the impact of picking and placing objects, the influence of muscle weakness, and the application of alternative data classification algorithms are essential to further enhance understanding of the interaction with objects and result in more accurate predictions.

The combination of the high number and the consequences of falls in older adults led to the development of fall risk assessments; non-sensor-based and sensor-based. Multiple studies used ML for older adults' fall risk prediction using raw IMU data. This study's objective was to develop a DL algorithm that predicts the fall risk of people living in a geriatric rehabilitation department using raw data collected from IMUs positioned at the ankles during the 10-m walk test.
Raw IMU data of 97 participants were used. The participants were classified as low, increased or high fall risk based on the Performance Oriented Mobility Assessment (POMA). Accelerometer and gyroscope's resultant time-series sequences (n=1037) were used as input for the Convolutional Neural Network (CNN) that was optimised and trained with 80% and tested with 20% of the participants. The results were compared with the performance of an existing portable sensor-based fall risk assessment called the Smart Floor (SF). The macro F1 of the unweighted (40%) and weighted (41%) multiclass classification CNNs was lower than the macro F1 of the SF (49%). The binary classification CNN's macro F1 (56%) was slightly lower than the SF's performance. All CNNs were better at predicting high-risk sequences. All models had poor performance when all three POMA fall risk categories should have been predicted. Adjustments to the data collection and CNN optimisation methods should be performed to study the possibility of predicting fall risk using raw IMU data in geriatric rehabilitation centres.

Goal: Build and evaluate the first prototype of the Cloud Walker, a passive assistive gait orthosis. Method: The Cloud Walker uses muscles of the upper body to store energy in the springs at the back of the Cloud Walker. This energy can than be used to swing the legs forward. The Cloud Walker was evaluated by 9 healthy participants walking on a treadmill. The participants walked at different speeds with and without the Cloud Walker. Finishing speed, EMG, heart rate, stride length and questionnaires are used to obtain insight on the performance of the Cloud Walker. Results: 3 out of 9 participants finished the test at the maximum speed of 4 km/h. The muscle activity of the measured upper leg muscles was higher with the Cloud Walker than without it. The heart rate increased more when walking with the Cloud Walker than walking without the Cloud Walker. The stride length at a slower pace was larger with the Cloud Walker. The stride length was smaller at a higher pace with the Cloud Walker than without it. The questionnaires showed that the Cloud Walker is a little uncomfortable and that walking with it takes a lot of effort. Discussion: It is expected that SCI patients are able to walk in the Cloud Walker since the device is comparable to other orthotic devices such as the ARGO. The Cloud Walker can obtain a higher speed and costs less effort than comparable devices.

Background: Radiostereometric analysis (RSA) is a technique to assess early migration, which has a predictive value for early failure of unicompartmental knee arthroplasty (UKA). To successfully analyze the RSA images it is important to have a sufficient amount of visiblemarkers with an acceptable three-dimensional distribution.
Objective: To provide insights on marker location distribution by developing a method to analyse marker placement based on visibility and accessibility.
Method: Marker locations in RSA scenes of patients (N=25) from a previous study are used to assess marker placement and analyze marker density by performing a voxelization operation to divide the bone in a grid of 3x3x3mm
bins. After spatial smoothing the hotspots with high marker fractions are determined and the four non-adjacent locations with the highest marker fraction are selected as the most favorable locations. Of these locations the distribution is analyzed by calculating the condition number (CN) of 1000 samples.
Results: The locations of a total of 635 3D markers in the femur and 917 3D markers in the tibia were assessed. After voxelization, the average marker density of non-empty bins was respectively 3.1 and 4.3 markers per bin. The four selected locations had a 47 times higher marker fraction compared to the average bin in the femur and 66 times higher for the selected locations in the tibia. Their mean condition number was 49.1 for the femur and 68.2 for the tibia.
Conclusion: The proposed method has proven to be sufficient for marker
placement analysis. Placing the tantalum markers in the selected locations, which have a considerably higher marker contribution to the total amount of
markers, would presumably increase the marker visibility during RSA and decrease patient exclusion due to an insufficient amount and/or distribution of visible markers.

To better understand and predict osteoarthritis, researchers are developing so-called coupled modelling workflows. Coupled workflows convert data from gait analysis studies to subject-specific tissue mechanical response estimations through the use of musculoskeletal and finite element models. The tissue mechanical response inside the joint is thought to play an integral role in the onset and progression of osteoarthritis. To study this, the design of coupled workflow was proposed. This design contained subject-specific gait data which was processed by a musculoskeletal model with a single degree of freedom knee joint. Musculoskeletal output was transferred through an adjusted generic finite element model of the knee to calculate maximum principal stress and shear strain values in the tibial cartilage of the knee. Proper marker placement is crucial for making an accurate assessment of the patient’s function in gait analysis studies. It has been claimed that marker misplacement is the main cause of measurement variability in gait analysis studies. It however remains unclear how potential marker misplacement propagates to the coupled modelling workflow results. To investigate this, in addition to the design of a coupled workflow, a sensitivity analysis was performed. With this sensitivity analysis we tried to answer the following question: How does marker placement of knee joint markers in gait analysis influence the tissue mechanical response calculated by a coupled workflow? For the sensitivity analysis, knee joint marker placements were virtually perturbed along anterior-posterior, proximal-distal and medial-lateral direction, to mimic marker misplacement. Corresponding knee biomechanics were estimated from the perturbed input data in the coupled workflow. Peak maximum principal stress values varied by up to 0.60MPa and peak shear strain varied by up to 0.08% as a result of perturbed knee marker placement. For cumulative stress levels, broader relative ranges were found. Moreover, the results showed that marker placement along the anterior-posterior direction had the greatest influence on corresponding tissue mechanical response estimations. In future studies a standard error of measurement margin is proposed in the assessment of coupled modelling worfklow results. In conclusion, the proposed workfow was relatively easy to build and provided similar tissue mechanical response result to those as reported by more complex models which were more computationally intensive. This implies that in the future, coupled modelling processes may very well be incorporated into the clinical decision-making process for musculoskeletal disorders like osteoarthritis.

Introduction: Revision total hip arthroplasty (THA) can be challenging in patients with severe acetabular bone defects and there is no consensus on the preferred reconstruction technique in this population. The aim of this study is to determine the biomechanical differences between Trabecular Metal Acetabular Revision System (TMARS) and Custom Triflange Acetabular Component (CTAC), in patients with Paprosky grade IIIB acetabular bone defects.
Methods: A retrospective study with patients undergoing revision THA surgery TMARS was conducted. For all these patients a virtual CTAC was designed. With this, the TMARS reconstruction could be compared with the virtual CTAC. Assessments of both implants included cortical and cancellous bone contact, center of rotation (COR), inclination, anteversion, and reamed bone volume.
Results: The mean bone contact was 5760 mm2 vs 5447 mm2 for TMARS and CTAC respectively. The TMARS showed significantly more cancellous bone contact than the CTAC (4966 mm2 vs 2887 mm2) and less cortical contact (795 mm2 vs 2560 mm2). The TMARS COR was significantly deviated in the medial-lateral direction compared to the CTAC, with a mean difference of 7.4 mm in lateral direction. TMARS showed a significantly greater reamed bone volume with a mean of 15997 mm3 for TMARS and 2292 mm3 for CTAC.
Conclusion: A semi-automatic method was designed for the determination of the biomechanical differences between TMARS and CTAC reconstructions. TMARS showed significant more cancellous bone contact, but also significant greater reamed bone volume with respect to CTAC. Furthermore, the TMARS COR was localized significantly more lateral compared to CTAC.

Robots can be powerful tools in post-stroke gait rehabilitation. However, state-of-the-art robots are often expensive machines containing rigid links with high inertia. Their expensiveness could limit their availability, and their high inertia reduces transparency, which could hinder rehabilitation progress. These factors raise the need for a minimalistic transparent robot that can effectively fill this gap. This research aims to design and validate such a device, a minimally actuated tendon-based exercise environment. The device is synthesized using an optimization algorithm that considers possible system configurations and optimizes both these configurations and their respective design parameters. Validation is done based on a reconstructed simulation of gait using motion capture on the optimal design to check whether it could be used for real-life rehabilitation. It was found that the most simplistic solution is not yet adequate for rehabilitation; thus, a slightly more complex design is required. While not providing the final solution, this research provides an important stepping stone towards designing a minimally actuated, simplistic, and transparent rehabilitation device.

Purpose: Patient-specific implants (PSIs) and surgical guides are widely used nowadays for the reconstruction of craniofacial defects. In this research, we quantitatively compared the realised position of craniofacial PSIs with the planned position and we calculated the difference between the realised osteotomy and the planned osteotomy if a resection preceded the reconstruction. Methods: We retrospectively included patients who received a craniofacial polyetheretherketone (PEEK) PSI in the LUMC between February 2019 and January 2021 and had a postoperative CT scan available. Postoperative CT scans were aligned with preoperative CT scans, followed by segmentation of the postoperative PSI position and skull, on which the realised osteotomy was defined by an inner and outer curve. Translation and rotation between the realised PSI position and planned PSI position were calculated as well as the surface distance between the realised osteotomy curves and the planned osteotomy. Results: 19 patients were included, regarding cranial (n = 11) and orbital (n = 8) implants. The translation vector length ranged from 0.5 mm to 6.7 mm and rotational deviation ranged from 0.9° to 17.4°, both being higher for orbital implants than for cranial implants (U = 3.00, p = .000, U = 15.0, p = .016). 12 patients were included in the osteotomy analysis. Mean distance between realised and planned osteotomy was 1.4 mm (SD 1.6) for outer curves and 0.5 mm (SD 1.7) for inner curves. Absolute mean distances over the two curves ranged from 0.6 mm (SD 0.5) to 4.1 mm (SD 2.6), accurately reproducing the planned osteotomy in most cases. Conclusion: In this study, there was a large variation in positioning accuracy for craniofacial PSIs. Cranial implants, being in good agreement with the planned PSI positions, were positioned with higher accuracy than orbital implants. In general, realised osteotomies were larger than planned. Better use and positioning of surgical guides could increase the PSI positioning accuracy. Clinical and aesthetical outcomes need to be included in future studies.

Vertebral fractures are the most common osteoporotic fractures, with a prevalence of 12-20% in Europe, making them a major health problem because of the associated morbidity, mortality and costs. Without adequate treatment, vertebral fractures are often followed by subsequent fractures, leading to further invalidation and deterioration of health. Therefore, early detection of vertebral fractures is important so preventative treatment can be initiated.

Vertebral fracture assessment (VFA) using Dual-Energy X-ray Absorptiometry (DXA) equipment is an imaging technique in which a lateral image of the spine is made. With vertebral morphometry, vertebral heights are measured and fractures are identified when vertebral height is lower than expected. However, vertebral morphometry is time and labor-intensive, and is subject to inter-operator variability. Automating VFA using Artificial Intelligence (AI) could help to overcome these limitations. We employed a co-development approach, aiming to create an AI-based tool to automatically perform vertebral morphometry on VFA images to identify vertebral fractures.

Firstly, we conducted a literature review to investigate the current use of AI in quantitative DXA imaging in a broad sense (Chapter 2). Besides VFA, other quantitative parameters describing bone macrogeometry and microgeometry can be extracted from DXA images. Incorporating these risk factors into multivariate prediction models could improve the identification of those at risk of fracture and help in clinical decision making. Although still in development, AI has been successfully applied in aid of fracture risk assessment, showing promising results.

In Chapter 3, the results of our reader study are described, evaluating VFA with manual vertebral morphometry as it is currently performed. This study served as a baseline measurement to quantify the effort needed to perform manual VFA and assess the potential value of automating VFA. The average annotation time per VFA image was 259 seconds. Although the intraclass correlation for vertebral height measurements between different readers was high, inter-observer agreement for fracture classification was only poor to moderate.

Together with our industry partner, we developed an AI-based software tool to perform vertebral morphometry. This tool is still in development, and we evaluated its current performance and potential impact in the study described in Chapter 4. Although its current standalone performance is suboptimal and shows room for improvement, this initial investigation showed that automated VFA has the potential to significantly reduce the required reader time.

Finally, in Chapter 5 we reflect on our user-centered co-development process and the next steps to bring our AI tool to market. We believe that collaboration between academic healthcare institutions and industry are essential for successful development of AI products. Validation of these products throughout the development process and actively involving intended users should be done. In the near future, vertebral fracture assessment can be supported by our AI-based application, potentially leading to lower annotation times and improving clinical workflow. However, further improvements have to be made and independent validation is required before market access.

Each year, more than 800 women in the Netherlands are diagnosed with cervical cancer, of which approximately half have locally advanced disease. The standard treatment for locally advanced cervical cancer (LACC) is external beam radiotherapy (EBRT) with concurrent chemotherapy followed by brachytherapy. Many patients experience some degree of treatment-related toxicity, mainly concerning the bowel, urinary tract, or vagina. Another important morbidity is hematologic toxicity (HT) due to bone marrow suppression, which might negatively impact the efficacy of adjuvant therapies. Treatment-related morbidity has a profound impact on patients’ quality of life. Proton therapy allows easier sparing of organs at risk (OARs), which might result in a decrease of treatment-related morbidities. One initiative to reduce treatment morbidities for LACC-patients was proposed in a collaborative project between Erasmus Medical Center (Erasmus MC), Leiden University Medical Center (LUMC), and Holland Proton Therapy Center (HollandPTC): the PROTECT-project. A clinical pilot study will be conducted to determine differences in dose to OARs and in morbidity outcomes when comparing state-of-the-art photon therapy with adaptive intensity-modulated proton therapy (aIMPT) for patients with LACC. Additionally, bone marrow sparing capabilities of both delivery techniques will be evaluated. The focus of this thesis is on the clinical implementation of aIMPT to facilitate the pilot-study and consists of three parts.

The thesis started with a systematic review of the literature about the relationship between bone marrow dose and HT in LACC-patients treated with primary chemoradiation. The review has been submitted for publication. There was a scarcity of studies investigating the relationship between bone marrow dose and HT and clinically useful prediction models were not available yet. The majority of the studies defining bone marrow as the whole pelvic bone found a significant association between bone marrow and HT, in contrast to studies evaluating lower density marrow spaces or active bone marrow. Future studies may use whole pelvic bone contouring to develop normal tissue complication probability models.

Secondly, the development and implementation of the treatment planning strategy for LACC-patients in HollandPTC were proposed. Uncertainties arising from proton therapy delivery were identified and strategies to address these uncertainties were determined. The proposed aIMPT-strategy for LACC-treatment consisted of a plan-of-the-day-approach with margins and robust planning. Further work includes establishing the balance between robustness settings and margins and optimizing the clinical implementation of the plan-of-the-day-strategy.

Lastly, the workflow for LACC-patients in Erasmus MC was translated into the HollandPTC environment. Requirements for the new workflow were generated with input from both the investigator and a risk evaluation with the users. A workflow was designed and translated into HollandPTC's situation. Secondly, the implementation of the plan-of-the-day-strategy in the treatment management process of the oncology information system ARIA was investigated. Recommendations were made to finalize and validate the implementation of the workflows.

This thesis provides a base for the clinical implementation of aIMPT for LACC in HollandPTC. The systematic review gives guidance for bone marrow sparing techniques in Erasmus MC and HollandPTC. Additionally, a combination of a plan-of-the-day-approach with margins and robust planning was identified as the most optimal treatment planning strategy. Lastly, an implementation strategy for the clinical workflow and treatment management process in HollandPTC was determined. Further work should focus on finalizing and validating the clinical implementation to facilitate PROTECT's clinical pilot study in HollandPTC.

Coronary artery disease (CAD) is one of the leading causes of death and disability worldwide. In CAD, the coronary arteries, that supply the myocardium with oxygen, are narrowed or even blocked by a process called atherosclerosis. Invasive coronary angiography (ICA) is the gold standard for the diagnosis of CAD, as well as for intraprocedural guidance of percutaneous coronary interventions. At this moment, stenosis severity is determined via visual inspection by a cardiologist. This method has several important drawbacks: a significant inter- and intra-rater variability and a high positive prediction bias. Currently available additional assessment techniques improve results, however, they come with prolonged procedural time, complication risks and increased costs. Additionally, a relevant number of syndromes exists that cannot be diagnosed sufficiently with these techniques. Automated software that uses all information within the ICA images and relates it to the context of complaints and outcomes, could be a valuable tool for improvement. Since machine learning (ML) can find relations between patient groups based on images, this could be a possible solution. However, despite promising results, clinical implementation of ML is still limited. Limited clinical applicability of developed algorithms and a lack of large high-quality datasets are the cause of this. In this research it was investigated if ML can help in solving the diagnostic and therapeutic challenges encountered in interventional cardiology (IC) and what is needed to apply ML to ICA images and other sources of coupled medical data. This was done in several steps. First the expectations and perceived barriers by interventional cardiologists on ML-based algorithms in clinical practice were assessed. Next, the feasibility of health insurance code-based querying of electronic health records (EHRs) for the creation of ML datasets was assessed. Third, a proof-of-concept study on creating a dataset and a deep learning network for predicting lesion significance on ICA images was carried out. Last, a roadmap for the curation of data for the development of ML models in IC was created.
Generally, interventional cardiologists have positive expectations for applying ML in their clinical practice. Furthermore, the willingness to collaborate in the development and clinical validation of ML algorithms is high. This is essential for translating ML models to clinical practice. Health insurance code-based querying of EHRs is not a feasible approach for creating datasets for the complex syndromes which currently pose diagnostic challenges in IC. However, the EHRs hold valuable data for the development of ML datasets. The same goes for the data collection strategy in the proof-of-concept study and it was shown that it is feasible to train deep learning networks on this data.
ML shows promising results for solving the diagnostic and therapeutic challenges encountered in interventional cardiology and directions for further research were identified. For the creation of algorithms that can be applied in clinical practice, close collaboration between ML professionals and clinicians is needed. Besides this, further research is needed to develop scalable strategies for the creation of large datasets, containing adequately labelled patients that represent the real-life population.

Introduction: Dorsal-Root-Ganglion (DRG) stimulation has been proposed as a neuromodulatory strategy in spinal cord injury (SCI) patients for the recovery of their impaired postural stability. A detailed evaluation of DRG stimulation-induced effects on postural stability can be obtained by recording center of pressure (CoP) and muscle activity during seated stability tasks, combined with the validated modified Functional Reach Test (mFRT). However, this particular combination of experiments has not yet been used to quantify postural stability in SCI patients or the healthy population. The aim of this study is to investigate postural stability in the healthy population utilizing a novel experimental protocol, involving this combination of experiments, to serve as reference for SCI patients in the upcoming long-term DRG stimulation study. Methods: Twenty healthy volunteers participated in the experiments. CoP sway during upright sitting and multi-directional leaning was determined from force plate recordings. Furthermore, functional arm reach was assessed with the mFRT. Lastly, contribution of hip and trunk muscles was examined by retrieving muscle activation patterns and intermuscular coherence from surface electromyography (EMG) recordings during multi-directional trunk movements. Results: During upright sitting an average root-mean-square CoP displacement of 1 mm, velocity of 4.5 mm/s and acceleration of 175 mm/s/s was found. Maximum CoP displacements during multi-directional leaning were within 10 to 20 cm. Furthermore, the average functional arm reach was 50 cm in the forward direction and 30 cm in the lateral directions. The contribution of hip and trunk muscles to multi-directional trunk movements consisted of antagonist and contralateral muscle activation, accompanied by high intermuscular synchronization between 0-100 Hz for most neighboring muscle pairs. Conclusion: This study provides a detailed examination of postural stability in the healthy population. Some revisions of the experimental protocol must be considered to reduce the amount of cross-registraton in EMG recordings. Nevertheless, the data of this study is well suited to be employed as reference for the potential improvements in postural stability in SCI patients during the upcoming long-term DRG stimulation study.