Comparison of Angle Measurements on Hallux Valgus with Two Different Methods Using Digital Images.
Members do not see these Ads. Sign Up.
Sung IH, Kim KC, Sung CH, Seo WY, Lee DY, Cho YA.
J Korean Foot Ankle Soc. 2013 Mar;17(1):40-44.
Tags:
-
-
Re: Measurement of Hallux Valgus Angle from Digital Images
Approach for measuring the angle of hallux valgus
Jin Zhou, Petr Hlavacek, Bo Xu, Wuyong Chen
Indian Journal of Orthopedics 2013 | Volume : 47 | Issue : 3 | Page : 278-282
-
Re: Measurement of Hallux Valgus Angle from Digital Images
Descriptive Quantitative Analysis of Hallux Abductovalgus Transverse Plane Radiographic Parameters.
Meyr AJ, Myers A, Pontious J.
J Foot Ankle Surg. 2013 Jun 17.
-
Re: Measurement of Hallux Valgus Angle from Digital Images
The hallux valgus angle of the margo medialis pedis as an alternative to the measurement of the metatarsophalangeal hallux valgus angle.
Klein C, Kinz W, Zembsch A, Groll-Knapp E, Kundi M.
BMC Musculoskelet Disord. 2014 Apr 21;15(1):133.
-
A comparison of hallux valgus angles assessed with computerised plantar pressure measurements, clinical examination and radiography in patients with diabetes
Daniël MC Janssen, Antal P Sanders, Nick A Guldemond, Joris Hermus, Gerard HIM Walenkamp and Lodewijk W van Rhijn
Journal of Foot and Ankle Research 2014, 7:33 doi:10.1186/1757-1146-7-33
-
Eric -
Hallux Valgus: Measurements and Characterization
F. Lamonacaa, M. Vasileb,A. Nastroc
Measurement; Available online 30 July 2014
Highlights
• Measurement methods for the objective and early detection of hallux valgus (HA) disease is proposed.
• The methods combine well known measurement techniques applied in engineering material science.
• The use of well-known techniques guaranties the traceability of the measurements.
• Samples from healthy and unhealthy bones with different HA severities are analyzed.
• Preliminary experimental tests asses the suitability and the effectiveness of the proposed methods.Click to expand...The aim of the paper is the characterization of a bone affected by hallux valgus pathology by measurement techniques normally used in material science. These techniques are able to show the differences between healthy bone and bone affected by hallux valgus. In particular, the ratio between organic and inorganic components of the bone is defined by measurement with thermogravimetric analysis. The different crystallinity of the bone samples are evaluated by X-ray powder diffraction techniques. The experimental results show that the proposed measurement techniques can be a prospective approach to study this type of pathology and to define the degree of the pathology.Click to expand... -
Anthropometric measurements are prone to to error - and it is incumbent upon the measurer to detail how they have dealt with the error associated with that measurement in a biological (not necessarily just statistically) valid manner. Measurement of angles in particular is prone to error of measurement as a lack of a standardised protocol. Years ago, I measured talar neck angles with an incredibly tight protocol via photographs - and could not get it better that +- about 3 degrees. The very slightest mis-presentation of the image will cause significant distortion of the angular value; the literature is heaving with this. Rob
Honorary Research Associate, Institute for Human Evolution, University of Witwatersrand
Fellow of The Centre For Human Biology, The University of Western Australia
"Please God, deliver me whole from my condemnation of those who criticise evolution......."
https://www.youtube.com/watch?v=uIwiPsgRrOs -
I have done more surgeries for HAV deformity than I can count and all of my anthropometric measurements have been done by eyesight only. Luckily, exact numbers are not necessary.
Steven -
Reliability of Two Smartphone Applications for Radiographic Measurements of Hallux Valgus Angles
Mauro Cesar Mattos e Dinato et al
The Journal of Foot and Ankle Surgery; Volume 56, Issue 2, March–April 2017, Pages 230–233
The objective of the present study was to assess the reliability of 2 smartphone applications compared with the traditional goniometer technique for measurement of radiographic angles in hallux valgus and the time required for analysis with the different methods. The radiographs of 31 patients (52 feet) with a diagnosis of hallux valgus were analyzed. Four observers, 2 with >10 years’ experience in foot and ankle surgery and 2 in-training surgeons, measured the hallux valgus angle and intermetatarsal angle using a manual goniometer technique and 2 smartphone applications (Hallux Angles and iPinPoint). The interobserver and intermethod reliability were estimated using intraclass correlation coefficients (ICCs), and the time required for measurement of the angles among the 3 methods was compared using the Friedman test. A very good or good interobserver reliability was found among the 4 observers measuring the hallux valgus angle and intermetatarsal angle using the goniometer (ICC 0.913 and 0.821, respectively) and iPinPoint (ICC 0.866 and 0.638, respectively). Using the Hallux Angles application, a very good interobserver reliability was found for measurements of the hallux valgus angle (ICC 0.962) and intermetatarsal angle (ICC 0.935) only among the more experienced observers. The time required for the measurements was significantly shorter for the measurements using both smartphone applications compared with the goniometer method. One smartphone application (iPinPoint) was reliable for measurements of the hallux valgus angles by either experienced or nonexperienced observers. The use of these tools might save time in the evaluation of radiographic angles in the hallux valgus.Click to expand... -
Automated Assessment of Hallux Valgus in Radiographic Images
Tomasz GąciarzWadim WojciechowskiZbisław Tabor
European Congress on Computational Methods in Applied Sciences and Engineering 13 October 2017
The purpose of the study was to develop an automated method for measurement of the selected angular variables, characterizing foot skeleton based on dorsoplantar projection radiographs. The study was a retrospective analysis of radiographic data. Totally, 50 dorsoplantar projection radiographs of a weight-wearing foot were analyzed (24 left and 26 right feet) of 32 patients (23 female, 9 male). Various quantities were measured to assess the severity of hallux valgus. The measurements were performed manually and with an automated method designed in the study. The automated and manual measurements were correlated. Repeated manual measurements were additionally performed to determine the variability of manual assessment of the hallux valgus. High correlation between manual and automated measurements have been observed. The accuracy of the framework is comparable with the accuracy of manual measurements.Click to expand... -
Footprint as an alternative to X-ray in hallux valgus angle measurement.
Pourhoseingholi E et al
Med J Islam Repub Iran. 2017 Jun 20;31:33. doi: 10.14196/mjiri.31.33.
Background: X- ray images provide accurate and reliable data in different foot pathologies. However, the accompanied complications will limit its use for epidemiological studies and research purposes, especially in children. Therefore, simple, accessible, and cost- effective methods such as footprint, with a good correlation with x-ray images, are needed to help diagnose different foot pathologies. In the present study, the accuracy of footprint technique in assessing hallux valgus angle (HVA) was evaluated based on x-ray images through measuring the angle between the medial border protrusion of the foot and the hallux. Methods: In this cross-sectional study, 42 participants with symptomatic hallux valgus were recruited. HVA was measured by both x-ray imaging and footprint. The differences between the two approaches were identified by applying correlation-coefficient test and reliability, which was assessed using interclass correlation (ICC). Results: A significant correlation was found between the HVA measured by x-ray and HVA by footprints (p< 0.001), and the ICC was upper than 90%. Conclusion: Foot print is a reliable method for measuring HVA, as it was highly correlated with the HVAs obtained by x-ray imaging.Click to expand... -
Non-Radiographic Measurement of Hallux Valgus Angle Using Self-Photography
Satoshi Yamaguchi et al
Journal of Orthopaedic & Sports Physical Therapy, 2018 Volume:0 Issue:0 Pages:1–23 DOI: 10.2519/jospt.2019.8280
Background
The reference standard of hallux valgus angle measurement is to use radiographs of the feet. However, its availability is constrained due to cost and radiation exposure. Less invasive non-radiographic assessments have been proposed. However, a measurement using self-photography has not been reported.
Objectives
To determine 1) the reliability of hallux valgus angle measurement using the same photographs of the feet (pHVA), 2) the reliability of repeated self-photography trials, and 3) the measurement error when the radiographic hallux valgus ankle (rHVA) was estimated using pHVA.
Methods
Participants took photographs of their own feet using a digital camera. The intra- and interrater reliability of pHVA measurements was assessed using intraclass correlation coefficient (ICC) and 95% minimum detectable change (MDC). The participants took photographs twice, and the reliability of repeated self-photography trials was examined. Participants also received radiographs of their feet from which the rHVA was measured. The measurement error was assessed using the mean difference and 95% limits of agreement.
Results
The intra- and interrater ICC of pHVA measurement was 0.98, with the MDCs less than 2°. The ICC of pHVA measurement for repeated self-photography was 0.96, and the MDC was 6.9°. pHVA was systematically lower than the rHVA by 5.3°.
Conclusion
Measurement of pHVA using self-photography was reproducible, although pHVA underestimated rHVA. pHVA can be a useful non-radiographic method to quantify hallux valgus deformity.Click to expand... -
New screening method for hallux valgus with using smartphone
Shibuya Hayatoshi, MD, Tomoyuki Nakasa, MD, Mikiya Sawa, MD, Yusuke Tsuyuguchi, MD, Munekazu Kanemitsu, MD, Yuki
Ota, MD, Nobuo Adachi, MD
AOFAS Annual Meeting 2018 1
Introduction/Purpose: We developed a new screening method for hallux valgus during weight bearing using the photocapturing
function in cell phones.
Methods: We compared apparent hallux valgus angles measured from photographs of people in a standing position during weight
bearing, which were captured on a smartphone (smartphone Hallux Valgus Angle: SP-HVA) (Smartphone Method), with hallux
valgus angles determined using conventional radiography in a standing posture during weight-bearing (X-ray HVA: XR-HVA). We
evaluated 37 feet of 25 patients (5 men and 25 women) who visited our hospital. The mean age at the time of the visit was 73.9
years. In the Smartphone Method, subjects stood upright and the examiner used a smartphone to take photos of both feet from
above while holding the phone at the level of the anterior knee.
Results: The mean value was 36.2° for SP-HVA and 36.7° for XR-HVA; the two methods showed a statistically significant
correlation (correlation coefficient of 0.86).
Conclusion: Our smartphone-assisted screening method for hallux valgus was possible to make a diagnosis equivalent to
conventional radiography method without radiation exposure.Click to expand... -
Intra- and Interobserver Agreement in Hallux Valgus Angle Measurements on Weightbearing and Non-Weightbearing Radiographs
Pietervan der Woude et al
The Journal of Foot and Ankle Surgery; 58, Issue 4, July 2019, Pages 706-712
The choice of treatment of hallux valgus deformity is influenced by angles measured on radiographs. Angles of interest are the hallux valgus angle (HVA), 1,2-intermetatarsal angle (IMA), and distal metatarsal articular angle (DMAA), as well as the presence of first metatarsophalangeal joint (MTPJ) subluxation. Guidelines for measuring those angles have been distributed by American Orthopaedic Foot and Ankle Society (AOFAS), although the influence of weightbearing on these angles and its clinical relevance is not clear. We conducted a study to determine the influence of weightbearing and the inter- and intraobserver agreement in the measurement. A total of 104 patients were enrolled in this study. Both weightbearing and non-weightbearing radiographs were obtained. In 2 rounds, 2 orthopedic surgeons and 2 musculoskeletal radiologists measured the angles in blinded digital radiographs according to AOFAS guidelines. Agreement on measurement of HVA, IMA, and DMAA in both weightbearing and non-weightbearing radiographs, as well as the presence of MTPJ subluxation, was calculated using the linear-weighted kappa coefficient and the intraclass correlation coefficient (ICC). Examiner agreement strength was defined according to the guidelines of Landis and Koch. HVA decreases significantly with weightbearing, whereas IMA significantly increases. The change in magnitude was 1° to 2° on average. No significant influence on DMAA could be noted. Interobserver agreement was excellent in both weightbearing and non-weightbearing radiographs for HVA (ICC 0.99 and ICC 0.99, respectively), IMA (ICC 0.98 and ICC 0.86, respectively), and DMAA (ICC 0.95 and ICC 0.97, respectively). The agreement on presence of subluxation was moderate to good (Fleiss kappa 0.50 to 0.63). Weightbearing alters forefoot geometry significantly. Adhering to AOFAS guidelines yields excellent interobserver agreement on HVA, IMA, and DMAA. First MTPJ subluxation presence is not an alternative for DMAA. The magnitude of change in IMA and HVA is small and therefore not clinically important. Both weightbearing and non-weightbearing radiographs can be used for determination of the correct treatment of hallux valgus deformity.Click to expand... -
The Role of Weightbearing Computed Tomography Scan in Hallux Valgus
Karim Mahmoud, MD, Sreenivasulu Metikala, MD, Samir D. Mehta, MD, ...
Foot & Ankle International November 4, 2020
Background:
Hyperpronation of the first metatarsal in hallux valgus (HV) is poorly understood by conventional weightbearing radiography. We aimed to evaluate this parameter using weightbearing computed tomography (WBCT) and to understand its association with other standard measurements.
Methods:
Retrospective evaluation of WBCT and weightbearing radiographs (WBXRs) was performed for 20 patients with HV feet and 20 controls with no such deformity. Axial computed tomography images of both groups were compared for the first metatarsal pronation angle (alpha angle) and tibial sesamoid subluxation (TSS) grades. The HV angle (HVA), first-second intermetatarsal angle (IMA), first metatarsal-medial cuneiform angle (MMCA), Meary’s angle, and calcaneal pitch (CP) angle of the study and control groups were compared on both WBXR and the corresponding 2-dimensional images of WBCT. All measurements were independently performed by 1 musculoskeletal radiology fellow and 1 foot and ankle surgical fellow. Measurements were averaged and interobserver reliability was calculated.
Results:
The HV group demonstrated significantly higher values for TSS grade (P < .001) but not for alpha angle (P = .121) compared with controls. Likewise, significantly elevated HVA and IMA were noted in the HV group on both imaging modalities, while no such differences were observed for the CP angle. Higher MMCA and Meary’s angle in the HV group were evident only on WBXR (MMCA, P = .039; Meary’s, P = .009) but not on WBCT (MMCA, P = .183; Meary’s, P = .171).
Among all, the receiver operating characteristic (ROC) curves demonstrated the greatest area under the curve (AUC) for HVA, followed by IMA. The alpha angle performed only just outside the range of chance (AUC, 0.65; 95% CI, 0.52-0.69). The Pearson’s correlations of the alpha angle, in the HV group, revealed a significant linear relationship with TSS grade and with HVA on WBXR, and only trended toward a weak linear relationship with IMA and with HVA on WBCT.
Conclusion:
The alpha angle, a measure of abnormal hyperpronation of the first metatarsal, was an independent factor that may coexist with other parameters in HV, but in isolation had limited diagnostic utility. “Abnormal” alpha angles were even observed in individuals without HV. Increases in IMA and MMCA were not necessarily associated with similar increases in alpha angle, despite moderate correlations with TSS grade and HVA on WBXR. Nevertheless, the WBCT was a useful method for assessing hyperpronation and guiding surgical management in individual cases.Click to expand... -
Evaluation of a Weightbearing CT Artificial Intelligence-Based Automatic Measurement for the M1-M2 Intermetatarsal Angle in Hallux Valgus
Jonathan Day, MS, Cesar de Cesar Netto, MD, PhD, Martinus Richter, MD, PhD, ...
Foot & Ankle International June 4, 2021
Background:
Weightbearing cone beam computed tomography (WBCT) has been gaining traction as a useful imaging modality in the diagnosis and follow-up of foot and ankle musculoskeletal pathologies due to the ability to perform quick, low-dose, 3-dimensional (3D) scans. However, the resulting wealth of 3D data renders daily clinical use time-consuming. The aim of this study was to evaluate a new artificial intelligence (AI)–based automatic measurement for the M1-M2 intermetatarsal angle (IMA) in hallux valgus (HV). We hypothesized that automatic and manual measurements would have a strong correlation, and that the AI software would yield better reproducibility and would be faster compared with manual measurements.
Methods:
This was a multicenter retrospective comparative case-control study in which a total of 128 feet were included from 93 patients who underwent WBCT scans as part of their routine follow-up: 59 feet with symptomatic HV and 69 controls. The IMA was measured automatically using the AI software and manually on digitally reconstructed radiographs (DRRs). The AI software produced both an automatic 2D (auto 2D) and 3D (auto 3D) measurement.
Results:
There were strong intermethod correlations between the DRR IMA and the auto 2D (HV, r = 0.61; control, r = 0.60; all P < .0001) and auto 3D (HV, r = 0.63; control, r = 0.52; all P < .0001) measurements, respectively. The intrasoftware reproducibility was very close to 100%. Measurements took 23.6 ± 2.31 seconds and 14.5 ± 1.18 seconds, respectively, when taken manually on DRRs and automatically. Controls demonstrated a mean DRR IMA of 8.6 (95% CI, 8.1-9.1), mean auto 2D of 11.2 (95% CI, 10.7-11.7), and mean auto 3D IMA of 11.0 (95% CI, 10.5-11.5). The HV group demonstrated significantly increased IMA compared with controls (P < .0001), with a mean DRR IMA of 15.4 (95% CI, 14.8-16.1), mean auto 2D of 17.8 (95% CI, 17.2-18.4), and mean auto 3D IMA of 16.8 (95% CI, 16.8-17.4).
Conclusion:
Measurements generated by the WBCT AI-based automatic measurement system for IMA demonstrated strong correlations with manual measurements, with near-perfect reproducibility. Further developments are warranted in order to make this tool more usable in daily practice, particularly with respect to its use in the presence of hardware in the foot.Click to expand... -
A novel rapid measurement of hallux valgus parameters using the built-in photo edit function of smartphones
Tianji Huang, Lin Wang, Chao Lu, Weiyang Zhong, Zenghui Zhao & Xiaoji Luo
BMC Musculoskeletal Disorders volume 22, Article number: 716 (2021)
Objective
The objective of this study was to assess the accuracy and reliability of and time taken by a novel method using the built-in photo-edit function of smartphones compared with PACS in measuring hallux valgus parameters.
Methods
Seventy patients (124 ft) admitted to our hospital with a diagnosis of hallux valgus without previous surgical procedures were retrospectively reviewed. The foot radiographs of all the patients were extracted from PACS. The hallux valgus angle (HVA) and the first and second intermetatarsal angles (IMAs) were measured by PACS and by this novel method using the built-in photo-edit function of a smartphone. The results of these two methods were compared, and the accuracy and reliability were assessed between these two methods.
Results
The average parameters measured by PACS were as follows: HVA average: 37.43 ± 9.61°; IMA average: 13.37 ± 4.01°. The average parameters measured by smartphones were as follows: HVA average: 37.09 ± 9.52° and IMA average: 13.49 ± 3.91°. When compared by the independent-samples T test, the average parameters between PACS and smartphones were not significantly different (HVA PACS vs HVA smartphones: P = 0.776; IMA PACS vs IMA smartphones: P = 0.816). The variability of the HVA (F = 0.166, P = 0.992) and IMA (F = 0.215, P = 0.982) measurements was similar for the PACS and smartphones. The ICCs of the average parameters of four measurements of HVA and IMA between PACS and smartphones were 0.995 (0.991–0.997) and 0.970 (0.958–0.979), indicating that the two methods were highly correlated. For the smartphone measurement, the interobserver and intraobserver reliability was very good for HVA and IMA. The average measurement time of PACS was 25.41 ± 0.86 s, and the average measurement time of smartphones was 20.29 ± 1.22 s. The smartphone time was significantly faster than that of PACS by approximately 5 s (P<0.001).
Conclusion
This novel method using the built-in photo-edit function of smartphones is accurate, reliable, convenient and time-saving in measuring the angles of hallux valgus.Click to expand... -
Development of a Manual Measurement Device for Measuring Hallux Valgus Angle in Patients with Hallux Valgus
Guoli Li et al
Int J Environ Res Public Health. 2022 Jul 26;19(15):9108
Background: Hallux valgus (HV) is one of the most common forefoot deformities, and its prevalence increases with age. HV has been associated with poor foot function, difficulty in fitting footwear and poor health-related quality of life. The aims of this study were to design and develop an easy-to-use measurement device for measuring hallux valgus angle (HVA) in patients with HV and to assess the measurement reliability of the newly designed measurement device.
Methods: A manual measurement device for measuring HVA was designed and developed to test on patients with HV. Two measuring methods, i.e., test-retest and intra-observer measurements, were used to evaluate the repeatability and reliability of the newly designed measurement device. In the test-retest measurements, a total of 42 feet from 26 patients with HV were repeatedly measured by the same researcher using the manual measurement device every 3 weeks over a period of 12 months. The measurement reliability of the newly designed measurement device was analysed based on the collected HVA data. In the intra-observer measurements, a total of 22 feet from the same group of HV patients were measured by the same researcher using the manual measurement device and by a consultant using X-ray measurement for comparison. The intraclass correlation coefficient (ICC) was used to determine the correlation of measurements between the manual measurement device and X-ray measurement.
Results: The mean of the difference between the two repeat measurements of HVA using the newly designed manual device was 0.62°, and the average of ICC was 0.995, which indicates excellent reliability. The ICC between X-ray and the average of twice-repeated manual measurements was 0.868, with 95% CI (0.649, 0.947) (p = 0.000). When the relationship in HVA between X-ray measurement and manual measurement using the new device was regressed as a linear relationship, the regression equation was y = 1.13x - 4.76 (R2 = 0.70).
Conclusions: The newly designed measurement device is easy to use, with low-cost and excellent reliability for HVA measurement, with the potential for use in clinical practice.Click to expand... -
Analysis of hallux valgus angles automatically extracted from 3D foot scans taken in North America, Europe and Asia
Yang Jiao, Sašo Džeroski, Jurca Ales
Ergonomics. 2022 Oct 21;1-13
The forefoot is the foot part most affected by ill-fitting shoes. Footwear fitting considers the measurements of length, width and arch length. Toe shape has not yet been used in sizing feet and fitting shoes. This study aims to investigate the variation in toe shape, as measured by the hallux valgus angle. An automatic and reproducible hallux valgus angle measuring method using 3D foot scans with no palpation markers is proposed and applied to about half a million samples collected across North America, Europe and Asia. The measuring method is robust and can detect the medial contour along the proximal phalanx even in extreme cases. The hallux valgus angle has a normal distribution with long tails on both sides in the general population. Large dispersions of HVA values were observed for both genders, and in all 3 geographical regions.Click to expand... -
Validity and reliability of hallux valgus angle measurement on smartphone digital photographs
Albert Cakar et al
J Foot Ankle Res. 2023 Oct 16;16(1):70
Background: This prospective study aimed to test the reliability and validity of hallux valgus angle (HVA) measurement on smartphone digital photographs compared with the standard radiographic evaluation.
Methods: Twenty Seven female patients (45 feet) with forefoot deformity were evaluated with weight-bearing anteroposterior foot radiographs and smartphone photographs. Radiographic hallux valgus angle (rHVA) was measured on digital radiographs. Two different photographic HVA measurement methods were used. In the first, the longitudinal axes of the first metatarsal and proximal phalanx were determined, and the angle between these axes was measured (pHVA), similar to the radiographic method. In the other method, the angle of the margo medialis pedis was measured on the photograph (pMMP). Two independent observers performed all measurements twice on two different occasions. Reliability analysis was performed using the interclass correlation coefficient. Agreement between the measurements was tested using Bland-Altman analysis.
Results: The repeated rHVA, pHVA and pMMP measurements showed excellent intra and inter-observer reliability, with ICC values above 0.900. The mean rHVA, pHVA, and pMMP were statistically similar (p:0.929, 27.03°±8.7°, 27.11°±8.8° and 26.5°±9.0° respectively). The mean difference between the rHVA and pHVA was - 0.07°±5.1° (range, --9.67 to 9.56°), and the mean difference between the rHVA and pMMP was 0.53°±4.4° (range, -9.76° to 8.22°). There was a strong positive correlation between both photographic methods and radiographic measurements (rho = 0.809, p = 0.001 and rho = 0.872, p = 0.001). In the Bland Altman plot, the upper and lower LOAs (95%CI) ranged from - 10.11° to 9.93° for rHVA and pHVA, and from - 8.26° to 9.33° for rHVA and pMMP. Linear regression analysis showed a proportional bias for pHVA but not for the pMMP (p:0.010 versus p:0.633, respectively). The range of the mean difference (prediction interval) between the pMMP and rHVA was 17.59° and 20° for pHVA and rHVA. Simple linear regression showed that the rHVA was predicted by the following equation: rHVA = 4.73 + 0.84 × pMMP (r2 = 0.761, p < 0.001).
Conclusions: Although measuring HVA through smartphone photographs is reliable, it is not a valid prediction method.Click to expand... -
Automatic estimation of hallux valgus angle using deep neural network with axis-based annotation
Ryutaro Takeda et al
Skeletal Radiol. 2024 Mar 13
Objectives: We developed the deep neural network (DNN) model to automatically measure hallux valgus angle (HVA) and intermetatarsal angle (IMA) on foot radiographs. The objective is to assess the accuracy of the model by comparing to the manual measurement of foot and ankle surgeons.
Materials and methods: A DNN was developed to predict the bone axes of the first proximal phalanx and all metatarsals from the first to the fifth in foot radiographs. The dataset used for model development consisted of 1798 radiographs collected from a population-based cohort and patients at our foot and ankle clinic. The retrospective validation cohort comprised of 92 radiographs obtained from 92 consecutive patients visiting our foot and ankle clinic. The mean absolute error (MAE) between automatic measurements by the model and the median of manual measurements by three foot and ankle surgeons was compared to 3° using one-tailed t-test and was also compared to the inter-rater difference in manual measurements among the three surgeons using two-tailed paired t-test.
Results: The MAE for HVA was 1.3° (upper limit of 95% CI 1.6°), and this was significantly smaller than the inter-rater difference of 2.0 ± 0.2° among the surgeons, demonstrating the superior accuracy of the model. In contrast, the MAE for IMA was 0.8° (upper limit of 95% CI 1.0°) that showed no significant difference from the inter-rater difference of 1.0 ± 0.1° among the surgeons.
Conclusion: Our model demonstrated the ability to measure the HVA and IMA with an accuracy comparable to that of specialists.Click to expand...
Loading...
- Similar Threads - Measurement Hallux Valgus
-
- Replies:
- 0
- Views:
- 7,042
-
- Replies:
- 0
- Views:
- 475
-
- Replies:
- 1
- Views:
- 1,366
-
- Replies:
- 0
- Views:
- 5,670
-
- Replies:
- 3
- Views:
- 2,312
-
- Replies:
- 0
- Views:
- 1,103
-
- Replies:
- 1
- Views:
- 1,673