Evaluation of a Weightbearing CT Artificial Intelligence-Based Automatic Measurement for the M1-M2 Intermetatarsal Angle in Hallux Valgus

Automatic software-based 3D-angular measurement for weight-bearing CT (WBCT) is valid

BACKGROUND

The purpose of this study was to compare automatic software-based angular measurement (AM) with validated measurement by hand (MBH) regarding angle values and time spent for Weight-Bearing CT (WBCT) generated datasets.

METHODS

Five-hundred WBCT scans from different pathologies were included in the study. 1st - 2nd intermetatarsal angle, talo-1st metatarsal angle dorsoplantar and lateral, hindfoot angle, calcaneal pitch angle were measured and compared between MBH and AM.

RESULTS

The pathologies were ankle osteoarthritis/instability, n = 147 (29%); Haglund deformity/Achillodynia, n = 41 (8%); forefoot deformity, n = 108 (22%); Hallux rigidus, n = 37 (7%); flatfoot, n = 35 (7%); cavus foot, n = 10 (2%); osteoarthritis except ankle, n = 82 (16%). The angles did not differ between MBH and AM (each p > 0.36). The time spent for MBH / AM was 44.5 / 1 s on average per angle (p < .001).

CONCLUSIONS

AM provided angles which were not different from validated MBH and can be considered as a validated angle measurement method. The time spent was 97% lower for AM than for MBH.

LEVELS OF EVIDENCE

Level III.

Artificial intelligence in musculoskeletal imaging: realistic clinical applications in the next decade

This article will provide a perspective review of the most extensively investigated deep learning (DL) applications for musculoskeletal disease detection that have the best potential to translate into routine clinical practice over the next decade. Deep learning methods for detecting fractures, estimating pediatric bone age, calculating bone measurements such as lower extremity alignment and Cobb angle, and grading osteoarthritis on radiographs have been shown to have high diagnostic performance with many of these applications now commercially available for use in clinical practice. Many studies have also documented the feasibility of using DL methods for detecting joint pathology and characterizing bone tumors on magnetic resonance imaging (MRI). However, musculoskeletal disease detection on MRI is difficult as it requires multi-task, multi-class detection of complex abnormalities on multiple image slices with different tissue contrasts. The generalizability of DL methods for musculoskeletal disease detection on MRI is also challenging due to fluctuations in image quality caused by the wide variety of scanners and pulse sequences used in routine MRI protocols. The diagnostic performance of current DL methods for musculoskeletal disease detection must be further evaluated in well-designed prospective studies using large image datasets acquired at different institutions with different imaging parameters and imaging hardware before they can be fully implemented in clinical practice. Future studies must also investigate the true clinical benefits of current DL methods and determine whether they could enhance quality, reduce error rates, improve workflow, and decrease radiologist fatigue and burnout with all of this weighed against the costs.

Developing Machine Learning-Based Predictive Models for Hallux Valgus Recurrence Based on Measurements From Radiographs

BACKGROUND

Machine learning (ML) is increasingly used to predict the prognosis of numerous diseases. This retrospective analysis aimed to develop a prediction model using ML algorithms and to identify predictors associated with the recurrence of hallux valgus (HV) following surgery.

METHODS

A total of 198 symptomatic feet that underwent chevron osteotomy combined with a distal soft tissue procedure were enrolled and analyzed from 2 independent medical centers. The feet were grouped according to nonrecurrence or recurrence based on 1-year follow-up outcomes. Preoperative weightbearing radiographs and immediate postoperative nonweightbearing radiographs were obtained for each HV foot. Radiographic measurements (eg, HV angle and intermetatarsal angle) were acquired and used for ML model training. A total of 9 commonly used ML models were trained on the data obtained from one institute (108 feet), and tested on the other data set from another independent institute (90 feet) for external validation. Optimal feature sets for each model were identified based on a 2000-resample bootstrap-based internal validation via an exhaustive search. The performance of each model was then tested on the external validation set. The area under the curve (AUC), classification accuracy, sensitivity, and specificity of each model were calculated to evaluate the performance of each model.

RESULTS

The support vector machine (SVM) model showed the highest predictive accuracy compared to other methods, with an AUC of 0.88 and an accuracy of 75.6%. Preoperative hallux valgus angle, tibial sesamoid position, postoperative intermetatarsal angle, and postoperative tibial sesamoid position were identified as the most selected features by several ML models.

CONCLUSION

ML classifiers such as SVM could predict the recurrence of HV (an HVA >20 degrees) at a 1-year follow-up while identifying associated predictors in a multivariate manner. This study holds the potential for foot and ankle surgeons to effectively identify individuals at higher risk of HV recurrence postsurgery.

Development and validation of a fully automated tool to quantify 3D foot and ankle alignment using weight-bearing CT

INTRODUCTION

Foot and ankle alignment plays a pivotal role in human gait and posture. Traditional assessment methods, relying on 2D standing radiographs, present limitations in capturing the dynamic 3D nature of foot alignment during weight-bearing and are prone to observer error. This study aims to integrate weight-bearing CT (WBCT) imaging and advanced deep learning (DL) techniques to automate and enhance quantification of the 3D foot and ankle alignment.

METHODS

Thirty-two patients who underwent a WBCT of the foot and ankle were retrospectively included. After training and validation of a 3D nnU-Net model on 45 cases to automate the segmentation into bony models, 35 clinically relevant 3D measurements were automatically computed using a custom-made tool. Automated measurements were assessed for accuracy against manual measurements, while the latter were analyzed for inter-observer reliability.

RESULTS

DL-segmentation results showed a mean dice coefficient of 0.95 and mean Hausdorff distance of 1.41 mm. A good to excellent reliability and mean prediction error of under 2 degrees was found for all angles except the talonavicular coverage angle and distal metatarsal articular angle.

CONCLUSION

In summary, this study introduces a fully automated framework for quantifying foot and ankle alignment, showcasing reliability comparable to current clinical practice measurements. This operator-friendly and time-efficient tool holds promise for implementation in clinical settings, benefiting both radiologists and surgeons. Future studies are encouraged to assess the tool's impact on streamlining image assessment workflows in a clinical environment.

Diagnostic applications and benefits of weightbearing CT in the foot and ankle: A systematic review of clinical studies

BACKGROUND

Foot and ankle weightbearing CT (WBCT) imaging has emerged over the past decade. However, a systematic review of diagnostic applications has not been conducted so far.

METHOD

A systematic literature search was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines after Prospective Register of Systematic Reviews (PROSPERO) registration. Studies analyzing diagnostic applications of WBCT were included. Main exclusion criteria were: cadaveric specimens and simulated WBCT. The Methodological Index for Non-Randomized Studies (MINORS) was used for quality assessment.

RESULTS

A total of 78 studies were eligible for review. Diagnostic applications were identified in following anatomical area's: ankle (n = 14); hindfoot (n = 41); midfoot (n = 4); forefoot (n = 19). Diagnostic applications that could not be used on weightbearing radiographs (WBRX) were reported in 56/78 studies. The mean MINORS was 9.8/24 (range: 8-12).

CONCLUSION

Diagnostic applications of WBCT were most frequent in the hindfoot, but other areas are on the rise. Post-processing of images was the main benefit compared to WBRX based on a moderate quality of the identified studies.

How AI May Transform Musculoskeletal Imaging

While musculoskeletal imaging volumes are increasing, there is a relative shortage of subspecialized musculoskeletal radiologists to interpret the studies. Will artificial intelligence (AI) be the solution? For AI to be the solution, the wide implementation of AI-supported data acquisition methods in clinical practice requires establishing trusted and reliable results. This implementation will demand close collaboration between core AI researchers and clinical radiologists. Upon successful clinical implementation, a wide variety of AI-based tools can improve the musculoskeletal radiologist's workflow by triaging imaging examinations, helping with image interpretation, and decreasing the reporting time. Additional AI applications may also be helpful for business, education, and research purposes if successfully integrated into the daily practice of musculoskeletal radiology. The question is not whether AI will replace radiologists, but rather how musculoskeletal radiologists can take advantage of AI to enhance their expert capabilities.

Artificial intelligence in foot and ankle surgery: current concepts

The twenty-first century has proven that data are the new gold. Artificial intelligence (AI) driven technologies might potentially change the clinical practice in all medical specialities, including orthopedic surgery. AI has a broad spectrum of subcomponents, including machine learning, which consists of a subdivision called deep learning. AI has the potential to increase healthcare delivery, improve indications and interventions, and minimize errors. In orthopedic surgery. AI supports the surgeon in the evaluation of radiological images, training of surgical residents, and excellent performance of machine-assisted surgery. The AI algorithms improve the administrative and management processes of hospitals and clinics, electronic healthcare databases, monitoring the outcomes, and safety controls. AI models are being developed in nearly all orthopedic subspecialties, including arthroscopy, arthroplasty, tumor, spinal and pediatric surgery. The present study discusses current applications, limitations, and future prospective of AI in foot and ankle surgery.

Automated decision support for Hallux Valgus treatment options using anteroposterior foot radiographs

BACKGROUND

Assessment of the potential utility of deep learning with subsequent image analysis to automate the measurement of hallux valgus and intermetatarsal angles from radiographs to serve as a preoperative aid in establishing hallux valgus severity for clinical decision-making.

AIM

To investigate the accuracy of automated measurements of angles of hallux valgus from radiographs for further integration with the preoperative planning process.

METHODS

The data comprises 265 consecutive digital anteroposterior weightbearing foot radiographs. 181 radiographs were utilized for training (161) and validating (20) a U-Net neural network to achieve a mean Sørensen–Dice index > 97% on bone segmentation. 84 test radiographs were used for manual (computer assisted) and automated measurements of hallux valgus severity determined by hallux valgus (HVA) and intermetatarsal angles (IMA). The reliability of manual and computer-based measurements was calculated using the interclass correlation coefficient (ICC) and standard error of measurement (SEM). Inter- and intraobserver reliability coefficients were also compared. An operative treatment recommendation was then applied to compare results between automated and manual angle measurements.

RESULTS

Very high reliability was achieved for HVA and IMA between the manual measurements of three independent clinicians. For HVA, the ICC between manual measurements was 0.96-0.99. For IMA, ICC was 0.78-0.95. Comparing manual against automated computer measurement, the reliability was high as well. For HVA, absolute agreement ICC and consistency ICC were 0.97, and SEM was 0.32. For IMA, absolute agreement ICC was 0.75, consistency ICC was 0.89, and SEM was 0.21. Additionally, a strong correlation (0.80) was observed between our approach and traditional clinical adjudication for preoperative planning of hallux valgus, according to an operative treatment algorithm proposed by EFORT.

CONCLUSION

The proposed automated, artificial intelligence assisted determination of hallux valgus angles based on deep learning holds great potential as an accurate and efficient tool, with comparable accuracy to manual measurements by expert clinicians. Our approach can be effectively implemented in clinical practice to determine the angles of hallux valgus from radiographs, classify the deformity severity, streamline preoperative decision-making prior to corrective surgery.

Ankle Joint Bone Density Distribution Correlates with Overall 3-Dimensional Foot and Ankle Alignment

BACKGROUND

Altered stress distribution in the lower limb may impact bone mineral density (BMD) in the ankle bones. The purpose of the present study was to evaluate the spatial distribution of BMD with use of weight-bearing cone-beam computed tomography (WBCT). Our hypothesis was that BMD distribution would be even in normal hindfeet, increased medially in varus hindfeet, and increased laterally in valgus hindfeet.

METHODS

In this study, 27 normally aligned hindfeet were retrospectively compared with 27 valgus and 27 varus-aligned hindfeet. Age (p = 0.967), body mass index (p = 0.669), sex (p = 0.820), and side (p = 0.708) were similar in the 3 groups. Hindfoot alignment was quantified on the basis of WBCT data sets with use of multiple measurements. BMD was calculated with use of the mean Hounsfield unit (HU) value as a surrogate. The HU medial-to-lateral ratio (HUR), calculated from tibial and talar medial and lateral half-volumes, was the primary outcome of the study.

RESULTS

The 3 groups significantly differed (p < 0.001) in terms of tibial HUR (median, 0.91 [interquartile range (IQR), 0.75 to 0.98] in valgus hindfeet, 1 [IQR, 0.94 to 1.05] in normal hindfeet, and 1.04 [IQR, 0.99 to 1.1] in varus hindfeet) and talar HUR (0.74 [IQR, 0.50 to 0.80] in valgus hindfeet, 0.82 [IQR, 0.76 to 0.87] in normal hindfeet, and 0.92 [IQR, 0.86 to 1.05] in varus hindfeet). Linear regression showed that all hindfoot measurements significantly correlated with tibial and talar HUR (p < 0.001 for all). The mean HU values for normally-aligned hindfeet were 495.2 ± 110 (medial tibia), 495.6 ± 108.1 (lateral tibia), 368.9 ± 80.3 (medial talus), 448.2 ± 90.6 (lateral talus), and 686.7 ± 120.4 (fibula). The mean HU value for each compartment was not significantly different across groups.

CONCLUSIONS

Hindfoot alignment and medial-to-lateral BMD distribution were correlated. In varus hindfeet, an increased HU medial-to-lateral ratio was consistent with a greater medial bone density in the tibia and talus as compared with the lateral parts of these bones. In valgus hindfeet, a decreased ratio suggested greater bone density in the lateral as compared with the medial parts of both the tibia and the talus.

LEVEL OF EVIDENCE

Prognostic Level III . See Instructions for Authors for a complete description of levels of evidence.

The Emerging Role of Automation, Measurement Standardization, and Artificial Intelligence in Foot and Ankle Imaging: An Update

In the past few years, advances in clinical imaging in the realm of foot and ankle have been consequential and game changing. Improvements in the hardware aspects, together with the development of computer-assisted interpretation and intervention tools, have led to a noticeable improvement in the quality of health care for foot and ankle patients. Focusing on the mainstay imaging tools, including radiographs, computed tomography scans, and ultrasound, in this review study, the authors explored the literature for reports on the new achievements in improving the quality, accuracy, accessibility, and affordability of clinical imaging in foot and ankle.

Can Weight-Bearing Computed Tomography Be a Game-Changer in the Assessment of Ankle Sprain and Ankle Instability?

Ankle sprain and chronic lateral ankle instability are complex conditions and challenging to treat. Cone beam weight-bearing computed tomography is an innovative imaging modality that has gained popularity, with a body of literature reporting reduced radiation exposure and operating time, and shortened examination time and a decreased time interval between injury and diagnosis. In this article, we make clearer the advantages of this technology and encourage researchers to investigate the area, and clinicians to use it as a primary mode of investigation. We also present clinical cases provided by the authors to illustrate those possibilities using advanced imaging tools.

Biomechanical Insights Afforded by Shape Modeling in the Foot and Ankle

Advancements in volumetric imaging makes it possible to generate high-resolution three-dimensional reconstructions of bones in throughout the foot and ankle. The use of weightbearing computed tomography allows for the analysis of joint relationships in a consistent natural position that can be used for statistical shape modeling. Using statistical shape modeling, a population-based statistical model is created that can be used to compare mean bone shape morphology and identify anatomical modes of variation. A review is presented to highlight the current work using statistical shape modeling in the foot and ankle with a future view of the impact on clinical care.

Advancements in Artificial Intelligence for Foot and Ankle Surgery: A Systematic Review

Background

There has been a rapid increase in research applying artificial intelligence (AI) to various subspecialties of orthopaedic surgery, including foot and ankle surgery. The purpose of this systematic review is to (1) characterize the topics and objectives of studies using AI in foot and ankle surgery, (2) evaluate the performance of their models, and (3) evaluate their validity (internal or external validation).

Methods

A systematic literature review was conducted using PubMed/MEDLINE and Embase databases in December 2022. All studies that used AI or its subsets machine learning (ML) and deep learning (DL) in the setting of foot and ankle surgery relevant to orthopaedic surgeons were included. Studies were evaluated for their demographics, subject area, outcomes of interest, model(s) tested, model(s)' performance, and validity (internal or external).

Results

A total of 31 studies met inclusion criteria: 14 studies investigated AI for image interpretation, 13 studies investigated AI for clinical predictions, and 4 studies were grouped as "other." Studies commonly explored AI for ankle fractures, calcaneus fractures, hallux valgus, Achilles tendon pathologies, plantar fasciitis, and sports injuries. For studies reporting the area under the receiver operating characteristic curve (AUC), AUCs ranged from 0.64 (poor) to 0.99 (excellent). Two studies (6.45%) reported external validation.

Conclusion

Applications of AI in the field of foot and ankle surgery are expanding, particularly for image interpretation and clinical predictions. Current model performances range from poor to excellent, and most studies lack external validation, demonstrating a need for further research prior to deploying AI-based clinical applications.

Level of Evidence

Level III, retrospective cohort study.

Current understanding on artificial intelligence and machine learning in orthopaedics - A scoping review

Background

Artificial Intelligence (AI) has improved the way of looking at technological challenges. Today, we can afford to see many of the problems as just an input-output system rather than solving from the first principles. The field of Orthopaedics is not spared from this rapidly expanding technology. The recent surge in the use of AI can be attributed mainly to advancements in deep learning methodologies and computing resources. This review was conducted to draw an outline on the role of AI in orthopaedics.

Methods

We developed a search strategy and looked for articles on PubMed, Scopus, and EMBASE. A total of 40 articles were selected for this study, from tools for medical aid like imaging solutions, implant management, and robotic surgery to understanding scientific questions.

Results

A total of 40 studies have been included in this review. The role of AI in the various subspecialties such as arthroplasty, trauma, orthopaedic oncology, foot and ankle etc. have been discussed in detail.

Conclusion

AI has touched most of the aspects of Orthopaedics. The increase in technological literacy, data management plans, and hardware systems, amalgamated with the access to hand-held devices like mobiles, and electronic pads, augur well for the exciting times ahead in this field. We have discussed various technological breakthroughs in AI that have been able to perform in Orthopaedics, and also the limitations and the problem with the black-box approach of modern AI algorithms. We advocate for better interpretable algorithms which can help both the patients and surgeons alike.

Automatic software-based 3D-angular measurement for Weight-Bearing CT (WBCT) provides different angles than measurement by hand

BACKGROUND

Purpose of this study was to compare automatic software-based angular measurement (AM, Autometrics, Curvebeam, Warrington, PA, USA) with previously validated measurement by hand (MBH) regarding angle values and time spent for the investigator for Weight-Bearing CT (WBCT).

METHODS

Five-hundred bilateral WBCT scans (PedCAT, Curvebeam, Warrington, PA, USA) were included in the study. Five angles (1st - 2nd intermetatarsal angle, talo-metatarsal 1-angle (TMT) dorsoplantar and lateral projection, hindfoot angle, calcaneal pitch angle) were measured with MBH and AM on the foot/ankle (side with pathology). Angles and time spent of MBH and AM were compared (t-test, homoscedatic).

RESULTS

The specific pathologies were ankle osteoarthritis/instability, n = 147 (29%); Haglund deformity/Achillodynia, n = 41 (8%); forefoot deformity, n = 108 (22%); Hallux rigidus, n = 37 (7%); flatfoot, n = 35 (7%); cavus foot, n = 10 (2%); osteoarthritis except ankle, n = 82 (16%). The angles differed between MBH and AM (each p < 0.001) except the calcaneal pitch angle (p = 0.05). The time spent for MBH / AM was 44.5 ± 12 s / 1 ± 0 s on average per angle (p < 0.0011).

CONCLUSIONS

AM provided different angles as MBH and can currently not be considered as validated angle measurement method. The investigator time spent is 97% lower for AM (1 s per angle) than for MBH (44.5 s per angle). Cases with correct angles in combination with almost no time spent showed the real potential of AM. The AM system will have to become reliable (especially in diminishing positive and negative angle values as defined) and valid which has to be proven by planned studies in the future.

LEVEL OF EVIDENCE

Level III.

Semi-automatic software-based 3D-angular measurement for Weight-Bearing CT (WBCT) in the foot provides different angles than measurement by hand

BACKGROUND

The purpose of this study was to compare semi-automatic software-based angular measurement (SAM) with previously validated measurement by hand (MBH) regarding angle values and time spent for the investigator for Weight-Bearing CT (WBCT).

METHODS

In this retrospective comparative study, five-hundred bilateral WBCT scans (PedCAT, Curvebeam, Warrington, PA, USA) were included in the study. Five angles (1st - 2nd intermetatarsal angle (IM), talo-metatarsal 1-angle (TMT) dorsoplantar and lateral projection, hindfoot angle, calcaneal pitch angle) were measured with MBH and SAM (Bonelogic Ortho Foot and Ankle, Version 1.0.0-R, Disior Ltd, Helsinki, Finland) on the right/left foot/ankle. The angles and time spent of MBH and SAM were compared (t-test, homoscesdatic).

RESULTS

The angles differed between MBH and SAM (mean values MBH/SAM; IM, 9.1/13.0; TMT dorsoplantar, -3.4/8.2; TMT lateral. -6.4/-1.1; hindfoot angle, 4.6/21.6; calcaneal pitch angle, 20.5/20.1; each p < 0.001 except the calcaneal pitch angle, p = 0.35). The time spent for MBH / SAM was 44.5 ± 12 s / 12 ± 0 s on average per angle (p < 0.001).

CONCLUSIONS

SAM provided different angles as MBH (except calcaneal pitch angle) and can currently not be considered as validated angle measurement method (except calcaneal pitch angle). The investigator time spent is 73% lower for SAM (12 s per angle) than for MBH (44.5 s per angle). SAM might be an important step forward for 3D-angle measurement of WBCT when valid angles are provided.

A Case-Control Study of 3D vs 2D Weightbearing CT Measurements of the M1-M2 Intermetatarsal Angle in Hallux Valgus

BACKGROUND

Weightbearing computed tomography (WBCT) 3-dimensional measurements may be reliable in assessing hallux valgus (HV). The objective of this study was to compare 2D and 3D WBCT measurements of the M1-M2 intermetatarsal angle (IMA) in patients with HV and in healthy controls. We hypothesized that 2D and 3D IMA measurements would correlate and have similar reliability in both HV and controls.

METHODS

Retrospective multicenter comparative study included WBCT scans from 83 feet (41 HV, 42 controls). IMA was measured on digitally reconstructed radiographs (DRR-IMA). 3D angle (3D-IMA) and its projection on the weightbearing plane (2D-IMA) were calculated from 3D coordinates of the first and second metatarsals. Intraobserver reliability and intermethod correlations were calculated using intraclass correlation coefficients (ICCs).

RESULTS

Intraobserver reliability was very strong for DRR-IMA (0.95) and 3D-IMA (0.99). Intermethod correlation between the 3 modalities in HV patients ranged from moderate (DRR vs 2D, 0.48; DRR vs 3D, 0.48) to very strong (2D vs 3D, 0.91). Similarly, intermethod correlation in the control group ranged from moderate (DRR vs 2D, 0.56; DRR vs 3D, 0.60) to very strong (2D vs 3D, 0.92).

CONCLUSION

Measurements for IMA are similar using DRR, 3D and 2D projected angles, with very strong intraobserver reliability and moderate to very strong intermethod correlations. This is the first head-to-head comparison between these measurement modalities in HV. Further investigations are warranted before formulating guidelines for the clinical use of 3D angles.

LEVEL OF EVIDENCE

Level III, case-control study.

Assessment of Flatfoot Deformity Using Digitally Reconstructed Radiographs: Reliability and Comparison to Conventional Radiographs

BACKGROUND

Digitally reconstructed radiographs (DRRs) generated from weightbearing computed tomography (WBCT) may potentially substitute for weightbearing plain radiographs (XRs) but have not been clinically validated. This study aims to test the reliability of 6 radiographic parameters of progressive collapsing foot deformity (PCFD) as measured on DRR, to investigate whether DRR represents comparably to XR through the same measurements, and to compare agreement of DRR and XR measurements of a standardized arch height parameter with reference measurements made on WBCT.

METHODS

DRR generated from preoperative WBCT of 71 patients (72 feet) treated surgically for PCFD were retrospectively compared with preoperative weight-bearing XR after exclusion criteria were applied. Six radiographic measurements were performed, including Meary angle, calcaneal pitch (CPA), medial cuneiform height (MCH), AP talar-first metatarsal angle (T-1MT), talonavicular coverage (TNCA), and talar incongruency (TIA). Arch height was measured on XR, DRR, and WBCT using a validated, standardized, navicular-based index. Intraclass correlation coefficients assessed DRR intraobserver and interobserver reliability. Paired samples t tests tested differences between XR and DRR. Bland-Altman limits of agreement analysis compared DRR and XR agreement with WBCT measurements.

RESULTS

Measurements were within standard PCFD ranges on XR and DRR. All measurements demonstrated excellent intrarater reliability and good to excellent interrater agreement, consistent with previous literature on XR. No differences were found for Meary, CPA, or TNCA. Minor differences were observed for MCH, T-1MT, and TIA. DRR measurements demonstrated greater agreement with WBCT than XR measurements.

CONCLUSION

DRR from WBCT may be a promising substitute for XR in the clinical evaluation of PCFD. Radiographic measurements made on DRR demonstrated good to excellent reliability. Although small differences were found between XR and DRR for certain measurements, DRR more accurately represented medial arch anatomy compared to gold standard WBCT data than XR. If validated as a clinical substitute, DRR could eventually obviate XR where WBCT is available.

LEVEL OF EVIDENCE

Level III, retrospective cohort study.

Feasibility study of hallux valgus measurement with a deep convolutional neural network based on landmark detection

OBJECTIVE

To develop a deep learning algorithm based on automatic detection of landmarks that can be used to automatically calculate forefoot imaging parameters from radiographs and test its performance.

MATERIALS AND METHODS

A total of 1023 weight-bearing dorsoplantar (DP) radiographs were included. A total of 776 radiographs were used for training and verification of the model, and 247 radiographs were used for testing the performance of the model. The radiologists manually marked 18 landmarks on each image. By training our model to automatically label these landmarks, 4 imaging parameters commonly used for the diagnosis of hallux valgus could be measured, including the first-second intermetatarsal angle (IMA), hallux valgus angle (HVA), hallux interphalangeal angle (HIA), and distal metatarsal articular angle (DMAA). The reference standard was determined by the radiologists' measurements. The percentage of correct key points (PCK), intragroup correlation coefficient (ICC), Pearson correlation coefficient (r), root mean square error (RMSE), and mean absolute error (MAE) between the predicted value of the model and the reference standard were calculated. The Bland-Altman plot shows the mean difference and 95% LoA.

RESULTS

The PCK was 84-99% at the 3-mm threshold. The correlation between the observed and predicted values of the four angles was high (ICC: 0.89-0.96, r: 0.81-0.97, RMSE: 3.76-6.77, MAE: 3.22-5.52). However, there was a systematic error between the model predicted value and the reference standard (the mean difference ranged from - 3.00 to - 5.08°, and the standard deviation ranged from 2.25 to 4.47°).

CONCLUSION

Our model can accurately identify landmarks, but there is a certain amount of error in the angle measurement, which needs further improvement.

Comparison between Weightbearing-CT semiautomatic and manual measurements in Hallux Valgus

BACKGROUND

Radiographic measurements are an essential tool to determine the appropriate surgical treatment and outcome for Hallux Valgus (HV). HV deformity is best evaluated by weight-bearing computed tomography (WBCT). The objective was (1) to assess the reliability of WBCT computer-assisted semi-automatic imaging measurements in HV, (2) to compare semi-automatic with manual measurements in the setting of an HV, and (3) to compare semi-automatic measurements between HV and control group.

METHODS

In this retrospective IRB (ID# 201904825) approved study, we assessed patients with hallux valgus deformity. The sample size calculation was based on the hallux valgus angle (HVA). Thus to obtain the 0.8 power, including 26 feet with HV in this study, was necessary. Our control group consisted of 19 feet from 19 patients without HV. Raw multiplanar data was evaluated using software CubeVue®. In the axial plane, hallux valgus angle (HVA), intermetatarsal angle (IMA), and interphalangeal angle (IPA) were measured. The semiautomatic 3D measurements were performed using the Bonelogic®Software. Inter-rater reliabilities were performed using ICC. Agreement between methods was tested using the Bland-Altman plots. The difference between Patologic and Control cases using semi-automatic measurements was assessed with the Wilcoxon signed-rank test. Alpha risk was set to 5% (α = 0.05). P ≤ 0.05 were considered significant.

RESULTS

Reliabilities utilizing ICC were over 0.80 for WBCT manual measurements and WBCT semi-automatic readings. Inter and intraobserver agreement for Manual and Semi-automatic WBCT measurements demonstrated excellent reliability.

CONCLUSIONS

Semi-automatic measurements are reproducible and comparable to measurements performed manually. The software differentiated pathological from non-pathological conditions when subjected to semi-automatic measurements. The development of advanced semi-automatic segmentation software with minimal user intervention is essential for the establishment of big data and can be integrated into clinical practice, facilitating decision-making.