scholarly journals Comparison of manual versus automated measurement of Cobb angle in idiopathic scoliosis based on a deep learning keypoint detection technology

2021 ◽  
Author(s):  
Yu Sun ◽  
Yaozhong Xing ◽  
Zian Zhao ◽  
Xianglong Meng ◽  
Gang Xu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Correlation Coefficient ◽  
X Rays ◽  
Automated Measurement ◽  
Keypoint Detection ◽  
Automated Method ◽  
Detection Technology ◽  
Short Period

Abstract Purpose The present study compared manual and automated measurement of Cobb angle in idiopathic scoliosis based on deep learning keypoint detection technology. Methods A total of 181 anterior–posterior spinal X-rays were included in this study, including 165 cases of idiopathic scoliosis and 16 normal adult cases without scoliosis. We labeled all images and randomly chose 145 as the training set and 36 as the test set. Two state-of-the-art deep learning object detection models based on convolutional neural networks were used in sequence to segment each vertebra and locate the vertebral corners. Cobb angles measured from the output of the models were compared to manual measurements performed by orthopedic experts. Results The mean Cobb angle in test cases was 27.4° ± 19.2° (range 0.00–91.00°) with manual measurements and 26.4° ± 18.9° (range 0.00–88.00°) with automated measurements. The automated method needed 4.45 s on average to measure each radiograph. The intra-class correlation coefficient (ICC) for the reliability of the automated measurement of the Cobb angle was 0.994. The Pearson correlation coefficient and mean absolute error between automated positioning and expert annotation were 0.990 and 2.2° ± 2.0°, respectively. The analytical result for the Spearman rank-order correlation was 0.984 (p < 0.001). Conclusion The automated measurement results agreed with the experts’ annotation and had a high degree of reliability when the Cobb angle did not exceed 90° and could locate multiple curves in the same scoliosis case simultaneously in a short period of time. Our results need to be verified in more cases in the future.

scholarly journals Automation analysis X-ray of the spine to objectify the assessment of the severity of scoliotic deformity in idiopathic scoliosis: a preliminary report

2020 ◽  
Vol 8 (3) ◽  
pp. 317-326
Author(s):  
Grigory A. Lein ◽  
Natalia S. Nechaeva ◽  
Gulnar М. Mammadova ◽  
Andrey A. Smirnov ◽  
Maxim M. Statsenko
Keyword(s):  
Neural Network ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Machine Learning Algorithms ◽  
Automated System ◽  
X Rays ◽  
X Ray ◽  
Scoliotic Deformity ◽  
Window Method ◽  
Treatment Type

Background. A large number of studies have focused on automating the process of measuring the Cobb angle. Although there is no practical tool to assist doctors with estimating the severity of the curvature of the spine and determine the best suitable treatment type. Aim. We aimed to examine the algorithms used for distinguishing vertebral column, vertebrae, and for building a tangent on the X-ray photographs. The superior algorithms should be implemented into the clinical practice as an instrument of automatic analysis of the spine X-rays in scoliosis patients. Materials and methods. A total of 300 digital X-rays of the spine of children with idiopathic scoliosis were gathered. The X-rays were manually ruled by a radiologist to determine the Cobb angle. This data was included into the main dataset used for training and validating the neural network. In addition, the Sliding Window Method algorithm was implemented and compared with the machine learning algorithms, proving it to be vastly superior in the context of this research. Results. This research can serve as the foundation for the future development of an automated system for analyzing spine X-rays. This system allows processing of a large amount of data for achieving 85% in training neural network to determine the Cobb angle. Conclusions. This research is the first step toward the development of a modern innovative product that uses artificial intelligence for distinguishing the different portions of the spine on 2D X-ray images for building the lines required to determine the Cobb angle.

Assessment of the plane of maximum curvature for patients with adolescent idiopathic scoliosis via computed tomography

2020 ◽  
Vol 44 (5) ◽  
pp. 298-304
Author(s):  
Hui-Dong Wu ◽  
Winnie Chiu-Wing Chu ◽  
Cheng-Qi He ◽  
Man-Sang Wong
Keyword(s):  
Computed Tomography ◽  
Adolescent Idiopathic Scoliosis ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Correlation Coefficient ◽  
Intraclass Correlation ◽  
Intrarater Reliability ◽  
Maximum Curvature ◽  
Curvature Measurements ◽  
The Mean

Background: In the assessment of three-dimensional features of adolescent idiopathic scoliosis, the plane of maximum curvature was compared with the coronal Cobb angle. Objectives: To investigate the intrarater reliability, variability, and difference of the prone plane of maximum curvature measurements taken from computed tomography using the constrained and unconstrained Cobb methods; to assess the difference and correlation between the prone plane of maximum curvature measurements obtained using the constrained and unconstrained Cobb methods; and to examine differences and correlation between the prone plane of maximum curvature Cobb angle and coronal Cobb angle measurements. Study design: Retrospective study. Methods: Records of 29 subjects with adolescent idiopathic scoliosis aged 15.8 ± 3.5 years were reviewed (25 thoracic and 24 thoracolumbar/lumbar curves). An experienced rater measured the plane of maximum curvature using the constrained and unconstrained Cobb methods, and the coronal Cobb angles using the conventional Cobb method on computed tomography images 3 times each with 1-week interval. The intraclass correlation coefficient (2,1), Pearson correlation coefficient ( r), one-way repeated measures analysis of variance, and paired t test were applied for various analyses. Results: The intraclass correlation coefficients for all intrarater reliability assessments were greater than 0.87. The plane of maximum curvature measurements of the two Cobb methods were excellently correlated ( r ⩾ 0.97) with no significant difference ( P > 0.05). The mean plane of maximum curvature Cobb angle was moderately correlated with ( r > 0.72) but significantly greater ( P < 0.001) than the mean coronal Cobb angle. Conclusion: The plane of maximum curvature measurements obtained from computed tomography were found to be reliable while the plane of maximum curvature measurements of the two Cobb methods were comparable. The mean plane of maximum curvature Cobb angle was moderately correlated with but significantly greater than the mean coronal Cobb angle. Clinical relevance The plane of maximum curvature measurements taken from computed tomography was found to be reliable, hence it could be used as a supplement to the coronal Cobb angle in the assessment and management of adolescent idiopathic scoliosis. With technological advancement, the radiation dose of computed tomography can be further reduced to a safer level for a broader range of cases.

scholarly journals May bending radiographs be replaced by recumbent CT scans in patients with adolescent idiopathic scoliosis?

2021 ◽  
Author(s):  
Max Prost ◽  
Joachim Windolf ◽  
Markus Rafael Konieczny
Keyword(s):  
Adolescent Idiopathic Scoliosis ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Strong Correlation ◽  
Ct Scans ◽  
X Rays ◽  
Exclusion Criteria ◽  
X Ray ◽  
A Minor ◽  
Main Curve

Abstract Purpose There is no data that show if it is possible to determine if a curve is structural or non-structural or to assess flexibility of an adolescent idiopathic scoliosis (AIS) by recumbent images like a CT scan (CTS) instead of bending radiographs (BR). We investigated if the results of BR may be compared to those of CTS. Methods We retrospectively analyzed prospectively collected data of patients with AIS in whom a selective spinal fusion was performed and in whom a CTS, BR, and full spine x-rays were made preoperatively. We measured the Cobb angles of the main and the minor curve in full spine x-ray, BR, and CTS. Results After applying inclusion and exclusion criteria, 39 patients were included. We found a strong correlation (r = 0.806, p < 0.01) between the Cobb angle of the main curve in BR and the Cobb angle of the main curve in the CTS and between the Cobb angle of the minor curve in BR and the Cobb angle of the minor curve in the CTS (r = 0.601, p < 0.01). All patients with a minor curve of less than 25 degrees in the BR had a Cobb angle of less than 35 degrees in the CTS. Conclusion Spinal curves showed a significant correlation between bending radiographs and recumbent images (CTS). In our group of patients, a Cobb angle of the minor curve of less than 35 degrees in the CTS indicated that this minor curve was non-structural.

scholarly journals Covid-19 Detection & Classification of chest X-rays using Deep Learning

2021 ◽  
Vol 9 (8) ◽  
pp. 2614-2619
Author(s):  
Aditya Singh
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Neural Network Model ◽  
X Rays ◽  
X Ray ◽  
Short Period ◽  
Medical Researchers ◽  
Entire World ◽  
Difficult Times

Abstract: The deadly Covid-19 virus, also known as the Coronavirus has affected the entire world in a short period of time. This pandemic has affected a lot of people in the entire world and caused many deaths. In these difficult times, it is important for the doctors and the medical researchers to differentiate accurately between positive cases and negative cases. This CNN (Convolutional Neural Network) model will allow us to classify X-ray images into positive cases and the normal ones. This dataset is collected from different public sources as well as from some hospitals and physicians. Our goal is to take help from these X- ray images and develop a model where it predicts and classifies the infected cases. Keywords: CNN, Prediction, Classification, Features, Training, Testing, Deep Learning

scholarly journals Idiopatinės paauglių skoliozės pooperaciniai rezultatai atliekant pedikulinę fiksaciją sraigtais

2013 ◽  
Vol 12 (4) ◽  
pp. 212-218
Author(s):  
Giedrius Bernotavičius ◽  
Kęstutis Saniukas ◽  
Vainius Rakauskas ◽  
Margarita Malceva
Keyword(s):  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Thoracic Spine ◽  
Screw Fixation ◽  
X Rays ◽  
Lateral Flexion ◽  
Operative Results ◽  
Lower Extremity Muscle ◽  
Pedicular Screw ◽  
Microsoft Office

Įvadas / tikslasDaugiasegmentė pedikulinė fiksacija sraigtais yra labiausiai paplitusi stuburo iškrypimo korekcijos metodika, nes naudojami pedikuliniai sraigtai didina korekcijos trijose plokštumose galimybes, trumpina fiksacijos ilgį, mažina torakoplastikos naudojimąir pooperacinį stuburo iškrypimo progresavimą. Šiuo metu sraigtai sukami į kiekvieną slankstelį priklausomai nuo pasirinkto fiksacijos lygio. Tačiau fiksacija sraigtais, literatūros duomenimis, padidina gydymo išlaidas 2–3 kartus, padidina komplikacijų skaičių. Pagrindinis šio tyrimo tikslas – išanalizuoti mūsų atliktos pedikulinės fiksacijos sraigtais pooperacinius rezultatus bei su sraigtais susijusias komplikacijas.Ligoniai ir metodaiAtliktas retrospektyvus tyrimas. Išanalizuoti 46 pacientai, kuriems 2010–2013 m. Vilniaus universiteto Santariškių vaikų ligoninės Vaikų ortopedijos-traumatologijos centro Ortopedijos skyriuje buvo atlikta nugarinė spondilodezė sraigtais. Tirti LenkeI tipo iškrypimą turintys paaugliai. Vertinti rentgenologiniai duomenys iš karto po operacijos, po 1 mėnesio, po 3 mėnesių ir 6 mėnesių. Darbe palyginti šoninio lenkimo duomenys prieš operaciją ir po operacijos. Išnagrinėta stuburo korekcija ir josrezultatų vidutinis pokytis per pusę metų bei komplikacijos. Statistinė analizė atlikta naudojantis statistiniu paketu SAS v9.0. Pagalbiniams skaičiavimams naudota „Microsoft Office Excel“ programa.RezultataiVidutinis krūtininės dalies iškrypimas prieš operaciją 50,37±12,99 Cobbo laipsnių, po operacijos – 17,61±10,14 Cobbo laipsnių. Krūtininės stuburo dalies korekcijos per pusę metų prarasta vidutiniškai tik 0,27º±4,52º, t. y. 1,53 %. Lyginant krūtininėsstuburo dalies šoninio lenkimo ir pooperacinius rezultatus gauta 38,85 % didesnė korekcija, nei buvo galima spėti iš šoninio lenkimo rentgenogramų (p=0,0001). Vidutiniškai dažniausiai atlikta Th4–L1 ilgio pedikulinė fiksacija sraigtais. Trims pacientamsnustatyta netaisyklinga sraigtų padėtis, tačiau atlikus kompiuterinės tomografijos tyrimą paaiškėjo, kad sraigtai nuo 2 mm iki 4 mm buvo perforavę pedikulinės kojytės vidinę sieną, tokia sraigtų padėtis buvo palikta ir papildomos operacijos neatliktos. Dviem pacientams koreguota sraigto padėtis pakartotinės operacijos metu. Šiems pacientams buvo ryškus skausminis sindromas, kojų raumenų jėgos silpnumas bei kojų hipestezijos. Dviem pacientams taikytas juosmeninis drenažas dėl likvorėjos po operacijos. Kraujagyslių, pleuros ar plaučių komplikacijų, susijusių su sraigtų padėtimi, nebuvo. Mūsų tyrime komplikacijos sudarė 8,6 %.IšvadosVidutinė atlikta krūtininės stuburo dalies korekcija yra 32,76º, t. y. vidutiniškai pakoreguota 65,04 % iškrypimo. Per pusę metų krūtininės stuburo dalies korekcijos prarasta vidutiniškai tik 1,53 %. Krūtininės stuburo dalies šoninio lenkimo matavimo rezultatai prieš operaciją ir po operacijos statistiškai reikšmingai skiriasi. Krūtininės stuburo dalies gauta korekcija 38,85 % didesnė, nei buvo galima numanyti iš šoninio lenkimo rentgenogramų. Skoliozės korekcija ir fiksacija sraigtais – saugi metodika.Reikšminiai žodžiai: idiopatinė skoliozė, nugarinė spondilodezė, pedikuliniai sraigtai.Postoperative results of posterior spondylodesis with pedicular screws in adolescent idiopathic scoliosis Background / objectiveFixation of multisegmental pedicular screw is the most commonly used spinal curvature correction technique for increasing the use of screws adjustment possibilities in three dimensions, shorten the length of fixation, reducing thoracoplasty use andpostoperative correction loss. Currently, srews are inserted in each vertebra depending on the level of fixation. However, this type of fixation has increased the cost of treatment 2–3 times and increased the number of complications. The main aim ofour study was to analyze our pedicular screw fixation postoperative results and related complications.Patients and methodsA retrospective study. Analysis of 46 patients from 2010 to 2013 years, Children’s Hospital, Affiliate of Vilnius University Hospital Santariskiu Klinikos pediatric orthopedics and traumatology center orthopedic department was performed for posteriorspondylodesis by screws. We analysed only the Lenke I scoliosis type. Data were collected from x-rays. Our data compare the lateral flexion (bending) and post-operative results. Analysing the average spinal adjustment, spinal correction was carriedout and the results mean change within half a year, and complications. Statistical analysis was performed using the statistical package SAS V9.0 and auxiliary calculations using the Microsoft Office Excel.ResultsThe average correction of thoracic spine before surgery was 50.37 º ± 12.99 º Cobb; the average curvature in thoracic spine after surgery was 17.61 ± 14.10 º Cobb angle. Thoracic spine had an average correction of a half-year loss of only 1.53%. Comparingthe thoracic spine lateral flexion (bending) and post-operative results, the obtained correction of 38.85 % is higher than it was possible to predict the side-bending x-rays (p = 0.0001). Three patients had screw malposition, but after the CT examination the violation was less than 4 mm. Two patients had reoperation for misplaced screws. These patients expressed a pain syndrome, lower extremity muscle strength and weakness of the leg, hypoaesthesia. Two patients received lumbar drainage for dural lesion after surgery. Cardiovascular, pulmonary or pleural complications related to screw position were not described. Complications in our study were 8.6 %.ConclusionsThe thoracic spine correction average was 32.76 ° Cobb angle, and the adjusted average was 65.04% of the curve. During the half-year loss, this part of the correction is minimal. Thoracic spine correction significantly different from that provided priorto the operation. Scoliosis correction and screw fixation is a safe method.Key words: idiopathic scoliosis, posterior spondylodesis, screw fixation

DEVELOPMENT AND EVALUATION OF AN AIRBAG BRACE SYSTEM FOR IDIOPATHIC SCOLIOSIS SUBJECTS

2014 ◽  
Vol 14 (02) ◽  
pp. 1450021
Author(s):  
WEN-CHIEH YANG ◽  
GERALD LIU ◽  
CHU-LUNG WU ◽  
KUO-CHIH SU ◽  
CHIH-HAN CHANG
Keyword(s):  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Lumbar Region ◽  
X Rays ◽  
Custom Made ◽  
Flexion Extension ◽  
Key Factor ◽  
Interface Parameters ◽  
Side Bending

There are disadvantages, especially from biomechanical viewpoint, to treating idiopathic scoliosis (IS) patients using traditional Boston braces with pressure pads. The major problem is the higher stiffness of the pressure pad as compared to the soft-tissue of the trunk. The aim of this study was to evaluate the effectiveness of Boston braces in which the pressure pad was replaced by the deformable airbag. A custom-made airbag brace which was designed based on the three-point compression principle of the Boston brace was prescribed to each of the 20 IS subjects with a mean age of 11.8 ± 2.8 years old. To evaluate the effectiveness of the treatment, the scoliosis Cobb angles before the treatment and after a follow-up period were measured from AP X-rays. The mean follow-up period was 11.5 ± 6.6 months. In addition, three interface parameters — contact pressure, contact area, and contact force — under seven postures — upright, flexion, extension, left side bending, right side bending, left side rotation and right side rotation — were measured using pressure mats to investigate interface conditions between airbags and skin. During the follow-up, all scoliosis curves showed Cobb angle reduction except one thoracic curve, which increased by one degree. The Cobb angle in the thoracic region decreased from 37.3 ± 10.6° to 27.3 ± 10.1° with a mean decreasing of 9.9 ± 5.9°. The Cobb angle in the lumbar region decreased from 30.1 ± 6.9° to 17.7 ± 8.6° with a mean decreasing of 12.4 ± 4.9°. Changes in the three interface parameters for the seven postures were insignificant. This indicated that the airbag brace could provide sufficient continuous corrective force. Using the airbag, instead of the pressure pad to provide the corrective force in the brace system can effectively correct the spinal curve caused by scoliosis. Easy adjustment and direct control of airbag pressure might be the key factor to the success of the airbag brace.

Upper instrumented vertebra to the right of the lowest instrumented vertebra as a predictor of an increase in the main thoracic curve after selective posterior fusion for the thoracolumbar/lumbar curve in Lenke type 5C adolescent idiopathic scoliosis: multicenter study on the relationship between fusion area and surgical outcome

2019 ◽  
Vol 31 (6) ◽  
pp. 857-864 ◽  
Author(s):  
Hiroki Oba ◽  
Jun Takahashi ◽  
Sho Kobayashi ◽  
Tetsuro Ohba ◽  
Shota Ikegami ◽  
...  
Keyword(s):  
Adolescent Idiopathic Scoliosis ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Posterior Fusion ◽  
Main Thoracic ◽  
Lowest Instrumented Vertebra ◽  
Line Tilt ◽  
The Right ◽  
Upper Instrumented Vertebra

OBJECTIVEUnfused main thoracic (MT) curvatures occasionally increase after selective thoracolumbar/lumbar (TL/L) fusion. This study sought to identify the predictors of an unacceptable increase in MT curve (UIMT) after selective posterior fusion (SPF) of the TL/L curve in patients with Lenke type 5C adolescent idiopathic scoliosis (AIS).METHODSForty-eight consecutive patients (44 females and 4 males, mean age 15.7 ± 2.5 years, range 13–24 years) with Lenke type 5C AIS who underwent SPF of the TL/L curve were analyzed. The novel “Shinshu line” (S-line) was defined as a line connecting the centers of the concave-side pedicles of the upper instrumented vertebra (UIV) and lowest instrumented vertebra (LIV) on preoperative radiographs. The authors established an S-line tilt to the right as S-line positive (S-line+, i.e., the UIV being to the right of the LIV) and compared S-line+ and S-line− groups for thoracic apical vertebral translation (T-AVT) and MT Cobb angle preoperatively, early postoperatively, and at final follow-up. The predictors for T-AVT > 20 mm at final follow-up were evaluated as well. T-AVT > 20 mm was defined as a UIMT.RESULTSAmong the 48 consecutively treated patients, 26 were S-line+ and 22 were S-line−. At preoperative, early postoperative, and final follow-up a minimum of 2 years later, the mean T-AVT was 12.8 mm (range −9.3 to 32.8 mm), 19.6 mm (range −13.0 to 41.0 mm), and 22.8 mm (range −1.9 to 68.7 mm) in the S-line+ group, and 10.8 mm (range −5.1 to 27.3 mm), 16.2 mm (range −11.7 to 42.1 mm), and 11.0 mm (range −6.3 to 26.9 mm) in the S-line− group, respectively. T-AVT in S-line+ patients was significantly larger than that in S-line− patients at the final follow-up. Multivariate analysis revealed S-line+ (odds ratio [OR] 23.8, p = 0.003) and preoperative MT Cobb angle (OR 7.9, p = 0.001) to be predictors of a UIMT.CONCLUSIONSS-line+ was defined as the UIV being to the right of the LIV. T-AVT in the S-line+ group was significantly larger than in the S-line− group at the final follow-up. S-line+ status and larger preoperative MT Cobb angle were independent predictors of a UIMT after SPF for the TL/L curve in patients with Lenke type 5C AIS. Surgeons should consider changing the UIV and/or LIV in patients exhibiting S-line+ during preoperative planning to avoid a possible increase in MT curve and revision surgery.

scholarly journals Sagittal profile modifications in hybrid versus all screw technique in adolescent idiopathic scoliosis

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Laura Scaramuzzo ◽  
Antonino Zagra ◽  
Giuseppe Barone ◽  
Stefano Muzzi ◽  
Leone Minoia ◽  
...  
Keyword(s):  
Adolescent Idiopathic Scoliosis ◽  
Idiopathic Scoliosis ◽  
Cobb Angle ◽  
Thoracic Kyphosis ◽  
Sagittal Alignment ◽  
Vertical Axis ◽  
Sagittal Profile ◽  
Sacral Slope ◽  
Sagittal Vertical Axis ◽  
Lenke Classification

AbstractAim of the study was to evaluate sagittal parameters modifications, with particular interest in thoracic kyphosis, in patients affected by adolescent idiopathic scoliosis (AIS) comparing hybrid and all-screws technique. From June 2010 to September 2018, 145 patients were enrolled. Evaluation included: Lenke classification, Risser scale, coronal Cobb angle, thoracic kyphosis (TK), lumbar lordosis (LL), sagittal vertical axis (SVA), pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS). Patients were divided in two groups (1 all-screws and 2 hybrid); a further division, in both groups, was done considering preoperative TK values. Descriptive and inferential statistical analysis was conducted. 99 patients were in group 1, 46 in group 2 (mean follow-up 3.7 years). Patients with a normo-kyphotic profile developed a little variation in TK (Δ pre–post = 2.4° versus − 2.0° respectively). Hyper-kyphotic subgroups had a tendency of restoring a good sagittal alignment. Hypo-kyphotic subgroups, patients treated with all-screw implants developed a higher increase in TK mean Cobb angle (Δ pre–post = 10°) than the hybrid subgroup (Δ pre–post = 5.4°) (p = 0.01). All-screws group showed better results in restoring sagittal alignment in all subgroups compared to hybrid groups, especially in hypo-TK subgroup, with the important advantage to give better correction on coronal plane.

Sign in / Sign up

Export Citation Format

Share Document