Monitoring of spine curvatures and posture during pregnancy using surface topography - case study and suggestion of method

Obtaining patient torso geometry for the design of scoliosis braces. A study of the accuracy and repeatability of handheld 3D scanners

OBJECTIVE

Obtaining patient geometry is crucial in scoliosis brace design for patients with adolescent idiopathic scoliosis. Advances in 3D scanning technologies provide the opportunity to obtain patient geometries quickly with fewer resources during the design process compared with the plaster-cast method. This study assesses the accuracy and repeatability of such technologies for this application.

METHODS

The accuracy and repeatability of three different handheld scanners and phone-photogrammetry was assessed using different mesh generation software. Twenty-four scans of a single subject's torso were analyzed for accuracy and repeatability based on anatomical landmark distances and surface deviation maps.

RESULTS

Mark II and Structure ST01 scanners showed maximum mean surface deviations of 1.74 ± 3.63 mm and 1.64 ± 3.06 mm, respectively. Deviations were lower for the Peel 1 scanner (maximum of -0.35 ± 2.8 mm) but higher with the use of phone-photogrammetry (maximum of -5.1 ± 4.8 mm). The mean absolute errors of anatomical landmark distance measurements from torso meshes obtained with the Peel 1, Mark II, and ST01 scanners were all within 9.3 mm (3.6%), whereas phone-photogrammetry errors were as high as 18 mm (7%).

CONCLUSIONS

Low-cost Mark II and ST01 scanners are recommended for obtaining torso geometries because of their accuracy and repeatability. Subject's breathing/movement affects the resultant geometry around the abdominal and anterolateral regions.

Validation and Assessment of a Posture Measurement System with Magneto-Inertial Measurement Units

Inappropriate posture and the presence of spinal disorders require specific monitoring systems. In clinical settings, posture evaluation is commonly performed with visual observation, electrogoniometers or motion capture systems (MoCaps). Developing a measurement system that can be easily used also in non-structured environments would be highly beneficial for accurate posture monitoring. This work proposes a system based on three magneto-inertial measurement units (MIMU), placed on the backs of seventeen volunteers on the T3, T12 and S1 vertebrae. The reference system used for validation is a stereophotogrammetric motion capture system. The volunteers performed forward bending and sit-to-stand tests. The measured variables for identifying the posture were the kyphosis and the lordosis angles, as well as the range of movement (ROM) of the body segments. The comparison between MIMU and MoCap provided a maximum RMSE of 5.6° for the kyphosis and the lordosis angles. The average lumbo-pelvic contribution during forward bending (41.8 ± 8.6%) and the average lumbar ROM during sit-to-stand (31.8 ± 9.8° for sitting down, 29.6 ± 7.6° for standing up) obtained with the MIMU system agree with the literature. In conclusion, the MIMU system, which is wearable, inexpensive and easy to set up in non-structured environments, has been demonstrated to be effective in posture evaluation.

Technical report: Rapid intraoperative reconstruction of cranial implants using additively manufactured moulds

During craniotomies, a portion of the calvarium or skull is removed to gain access to the intracranial space. When it is not possible to re-implant the flap, surgeons may repair the defect intraoperatively or at a later date. With larger defects being more difficult to repair intraoperatively, we investigated a method for the creation of patient-specific moulds for ad hoc bone flap reconstruction using rapid prototyping. Patient-specific moulds were created based on light scanned models of the defect, using custom software and rapid prototyping. Polymethylmethacrylate bone implants were created for three retrospective craniotomy cases and evaluated based on original flap and skull reconstruction accuracy. Bone implants created using our moulding method reconstruct the original flap and skull with an average reconstruction accuracy of 0.82 and 1.3 mm, respectively. Average skull reconstruction accuracy obtained by surgeons performing freehand implant reconstruction was 1.49 mm. Time needed to generate moulds was between 2 h and 45 min and 6 h and 20 min. Improvements to current printing technology will make this procedure technically feasible for future cranial procedures.

Influence of type 1 diabetes on the postural control of women in the third gestational trimester

BACKGROUND

Diabetes can cause biomechanical alterations that may be responsible for additional changes to those existing in a regular gestational period. The way a maternal body responds when affected by diabetes has not been clearly understood. This study aimed to describe the influence of type 1 diabetes on pregnant women's postural control.

METHODS

Forty pregnant women in their third gestational trimester were allocated in two equal groups - the control group and the type 1 diabetic group. The variables related to postural control and balance were assessed using photogrammetry (head protrusion; cervical lordosis; thoracic kyphosis; lumbar lordosis; pelvic anteversion, knee flexion, tibiotarsal and foot inclination angles were measured), and baropodometry (anteroposterior and mediolateral distance trajectory of the center of pressure, amplitude and average speed of displacement of the center of pressure).

FINDINGS

The results of the type 1 diabetic group showed, in the postural analysis, lower head protrusion and pelvic anteversion angles, while there was higher cervical lordosis, thoracic kyphosis, and lumbar lordosis angles. In the baropodometry, the anteroposterior distance and the amplitude of the center of pressure displacement with eyes open and closed were higher.

INTERPRETATION

The findings suggest that type 1 diabetes mellitus in the third trimester of pregnancy is associated with postural changes, a decrease in the active ankle range of motion and increase in the anteroposterior oscillation of the center of pressure, with negative repercussions for postural control.

Intraoperative integration of structured light scanning for automatic tissue classification: a feasibility study

PURPOSE

Structured light scanning is a promising inexpensive and accurate intraoperative imaging modality. Integration of these scanners in surgical workflows has the potential to enable rapid registration and augment preoperative imaging, in a practical and timely manner in the operating theatre. Previously, we have demonstrated the intraoperative feasibility of such scanners to capture anatomical surface information with high accuracy. The purpose of this study was to investigate the feasibility of automatically characterizing anatomical tissues from textural and spatial information captured by such scanners using machine learning. Assisted or automatic identification of relevant components of a captured scan is essential for effective integration of the technology in surgical workflow.

METHODS

During a clinical study, 3D surface scans for seven total knee arthroplasty patients were collected, and textural and spatial features for cartilage, bone, and ligament tissue were collected and annotated. These features were used to train and evaluate machine learning models. As part of our preliminary preparation, three fresh-frozen knee cadaver specimens were also used where 3D surface scans with texture information were collected during different dissection stages. The resulting models were manually segmented to isolate texture information for muscles, tendon, cartilage, and bone. This information, and detailed labels from dissections, provided an in-depth, finely annotated dataset for building machine learning classifiers.

RESULTS

For characterizing bone, cartilage, and ligament in the intraoperative surface models, random forest and neural network-based models achieved an accuracy of close to 80%, whereas an accuracy of close to 90% was obtained when only characterizing bone and cartilage. Average accuracy of 76-82% was reached for cadaver data in two-, three-, and four-class tissue separation.

CONCLUSIONS

The results of this project demonstrate the feasibility of machine learning methods to accurately classify multiple types of anatomical tissue. The ability to automatically characterize tissues in intraoperatively collected surface models would streamline the surgical workflow of using structured light scanners-paving the way to applications such as 3D documentation of surgery in addition to rapid registration and augmentation of preoperative imaging.

Decreased Vertical Trunk Inclination Angle and Pelvic Inclination as the Result of Mid-High-Heeled Footwear on Static Posture Parameters in Asymptomatic Young Adult Women

The influence of high-heel footwear on the lumbar lordosis angle, anterior pelvic tilt, and sacral tilt are inconsistently described in the literature. This study aimed to investigate the impact of medium-height heeled footwear on the static posture parameters of homogeneous young adult standing women. Heel geometry, data acquisition process, as well as data analysis and parameter extraction stage, were controlled. Seventy-six healthy young adult women with experience in wearing high-heeled shoes were enrolled. Data of fifty-three subjects were used for analysis due to exclusion criteria (scoliotic posture or missing measurement data). A custom structured light surface topography measurement system was used for posture parameters assessment. Three barefoot measurements were taken as a reference and tested for the reliability of the posture parameters. Two 30-degree wedges were used to imitate high-heel shoes to achieve a repeatable foot position. Our study confirmed the significant (p < 0.001) reduced vertical balance angle and pelvis inclination angle with large and medium-to-large effects, respectively, due to high-heel shoes. No significant differences were found in the kyphosis or lordosis angles. High-heeled shoes of medium height in young asymptomatic women can lead to a straightening effect associated with a reduced vertical balance angle and decreased pelvic inclination.

Accuracy of 3D surface scanners for clinical torso and spinal deformity assessment

Externally visible deformities are cosmetic features of great concern for Adolescent Idiopathic Scoliosis (AIS) patients. Current assessment techniques for AIS do not fully encompass the external deformity. A non-invasive method capable of capturing superficial anatomy, such as 3D scanning, would enable better qualitative and quantitative evaluation of cosmesis. This study aimed to quantify the accuracy of commonly available scanners, in assessing posterior asymmetry in AIS. The technique of 3D surface deviation analysis was proposed as a suitable method for comparing the models created by each scanner. Eight plaster cast moulds manufactured to create braces for AIS patients were used as test samples. Four 3D scanners were selected: Solutionix RexScan CS+; Artec Eva; Microsoft Kinect V1; iPhone with 123D Catch App. These scanners were selected from those available as representative of a range of scanning technologies. Each cast was scanned and 3D models created. A simulated rib hump measurement was obtained and the surface-to-surface deviations between the Solutionix scan and all other scans were determined. The Solutionix scanner is a metrology scanner of very high quality and so it was selected as the reference. Surface-to-surface deviations were calculated in the positive and negative directions separately to specifically identify size and volume inaccuracies created by the scans. Surface deviations showed excellent agreement between the Solutionix and the Eva with deviations of +0.17 ± 0.17 mm (Eva regions larger) and -0.20 ± 0.32 mm (Eva regions smaller) (mean±SD). The Kinect showed lower agreement (+1.58 ± 1.50 mm and -0.58 ± 0.58 mm). The iPhone scans were not able to be scaled to the correct size, so were excluded. Rib hump measurements with all scanners were within clinical measurement variability (±4.9 deg) of the known values. These commercially available 3D scanners are capable of imaging torso shape in 3D and deriving clinically relevant external deformity measures. The non-invasive 3D topographic information provided can be used to improve assessment of torso shape in spinal deformity patients.

The effect of a maternity support belt on static stability and posture in pregnant and non-pregnant women

PURPOSE

Physical and hormonal changes during pregnancy are thought to affect balance and injury risk, with increased numbers of falls being reported. A maternity support belt (MSB) has been suggested to stabilize the pelvis and to enhance balance. The purpose of this study was therefore to investigate the effect of an MSB on postural stability in different trimesters of pregnancy.

METHODS

Postural stability was assessed in the first (T1, n = 30), second (T2, n = 30) and third trimester (T3, n = 30) of pregnancy and compared to non-pregnant controls (n = 30), using a portable force plate. Postural sway during quiescent standing with and without applying an MSB was characterized by analyzing path length, velocity, amplitudes and area. Subsequently, anterior and posterior limits of stability (LoS) were determined.

RESULTS

Postural sway during quiescent standing did not change with pregnancy. However, LoS performance was reduced already in T1, before body mass significantly increased. The MSB led to a small improvement in the LoS while slightly increasing postural sway in anterior-posterior direction and shifting the center of pressure posteriorly during quiescent standing.

CONCLUSION

While impairments in balance already occurred early in pregnancy before body mass significantly increased, they were subtle and only measurable in exacerbated conditions. This challenges the assumed necessity of balance enhancing interventions in pregnant women. Although the MSB significantly affected body posture, the magnitude of the LoS improvement using the MSB was very small. Thus, it remains debatable if the MSB is a meaningful tool to increase balance during pregnancy.

Imaging in the Diagnosis and Monitoring of Children with Idiopathic Scoliosis

The paper reviews the current imaging methods in the diagnosis and monitoring of patients with adolescent idiopathic scoliosis. Radiography is generally used in the initial diagnosis of the condition. Postero-anterior erect full spine radiograph is generally prescribed, and is supplemented by lateral full spine radiograph when indicated. To reduce the radiation hazard, only the area of interest should be exposed, and follow-up radiographs should be taken with as few projections as possible. When available, EOS® stereoradiography should be used. The radiation of the microdose protocol is 45 times less than that of the conventional radiography. Surface topography offers another approach to monitoring changes of curvatures in AIS patients. Recently, 3D ultrasound has been found to be able to measure the Cobb angle accurately. Yet, it is still in the early developmental stages. The inherent intrinsic and external limitations of the imaging system need to be resolved before it can be widely used clinically. For AIS patients with atypical presentation, computed tomography (CT) and/or magnetic resonance imaging (MRI) may be required to assess for any underlying pathology. As CT is associated with a high radiation dose, it is playing a diminishing role in the management of scoliosis, and is replaced by MRI, which is also used for pre-operative planning of scoliosis. The different imaging methods have their limitations. The EOS® stereoradiography is expensive and is not commonly available. The surface topography does not enable measurement of Cobb angle, particularly when the patient is in-brace. The 3D ultrasound scanning has inherent intrinsic technical limitation and cannot be used in all subjects. Radiography, however, enables diagnosis and monitoring of the adolescent idiopathic scoliosis (AIS). It is thus the gold standard in the evaluation and management of scoliosis curves.