Font Size: a A A

Study On The Diagnostic Value Of Ultrasound Features And Models To Identify Benign And Malignant Orbital Neoplastic Lesions

Posted on:2024-06-23Degree:MasterType:Thesis
Country:ChinaCandidate:Y Y HuangFull Text:PDF
GTID:2544307175976409Subject:Imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Background:The orbit contains multiple tissue components such as lacrimal gland,conjunctiva,eyelid and orbital soft tissue.Orbital neoplastic lesions have the characteristics of multiple tissue sources,complex diseases,insidious growth locations,different growth patterns,and unclear clinical symptoms,which made it relatively difficult to identify benign and malignant orbital neoplastic lesions.Imaging examination can improve the diagnostic accuracy of orbital neoplastic lesions and is of great significance for clinical selection of the best treatment plan.Magnetic resonance imaging(MRI)has become a routine preoperative examination for orbital neoplastic lesions because of its excellent soft tissue resolution.Currently,the common examination sequences of orbital MRI include T1-weighted imaging(T1WI),T2-weighted imaging(T2WI)and fat suppression sequences(FS).The high spatial resolution and multi-directional imaging of MRI enables accurate localization of neoplastic lesions and facilitates zoning and morphological observation.Neoplastic lesions of different pathological types have certain differences in routine T1 WI,T2WI and enhancement.FS can determine whether the tumor contains lipid components which facilitate the identification of tumors with fat components.Thus,MRI is highly advantageous in identifying benign and malignant orbital neoplastic lesions.However,MRI is expensive,and MRI is not suitable for patients with metal-containing implants in the body and patients whose bodies cannot tolerate transport and movement.With the superficial location of the orbit and the high resolution of the ultrasound highfrequency probe on the superficial soft tissues,the neoplastic lesions can be clearly displayed.The capsular structure of the eye can serve as a good sound transmission window,making the image of the soft tissue also clearly obtainable.The real-time nature of ultrasound can be used to determine the relative motion of the lesion and understand the relationship between the lesion and the adjacent tissues through the patient’s eye movements such as turning the eyeball and blinking.Color doppler flow imaging(CDFI)and pulsed wave doppler(PWD)can be used to reflect the blood supply and hemodynamic characteristics of neoplastic lesions.Ultrasound can identify neoplastic lesions and effectively identify benign and malignant lesions based on 2D,CDFI,and PWD ultrasound imaging.The ultrasound features of orbital neoplastic lesions are mostly reported as single cases and rarely summarized,and some benign and malignant lesions are difficult to identify.Therefore,summarizing the ultrasound features of orbital neoplastic lesions,combining clinical characteristics to construct an ultrasonic diagnostic model to identify benign and malignant orbital neoplastic lesions will help grasp key points for distinguishing between benign and malignant orbital neoplastic lesions,improve diagnostic efficiency and accuracy.Objective:This study constructed an ultrasonic diagnostic model for distinguishing benign and malignant orbital neoplastic lesions by retrospectively analyzing the ultrasonic characteristics of 488 cases of orbital neoplastic lesions and combining clinical features.At the same time,the established ultrasonic diagnostic model was used to distinguish benign and malignant orbital neoplastic lesions,and the diagnostic efficacy of ultrasonic diagnostic model and MRI in distinguishing benign and malignant orbital neoplastic lesions were compared with pathological results as the gold standard.Materials and MethodsPart Ⅰ: Clinical and ultrasonic feature analysis of benign and malignant orbital neoplastic lesions and establishment of diagnostic model1.The clinical data and ultrasound images of 361 benign and 127 malignant orbital neoplastic lesions admitted to our hospital were collected,and the clinical and ultrasound characteristics of benign and malignant neoplastic lesions were summarized.2.The clinical manifestations and ultrasonographic features of benign and malignant orbital neoplastic lesions were screened by single-factor analysis.3.Multi-factor logistic regression analysis was used to construct a diagnostic model for differentiating benign and malignant orbital neoplastic lesions.4.The diagnostic efficacy of the diagnostic model in identifying benign and malignant orbital neoplastic lesions was analyzed by drawing receiver operating characteristic(ROC)curves.Part Ⅱ: A comparative study on the value of ultrasonic diagnostic model and magnetic resonance imaging diagnosis of common orbital neoplastic lesions1.The ultrasound and MRI images of 270 benign and 96 malignant orbital neoplastic lesions confirmed by surgical pathology and examined by both orbital ultrasound and MRI in our hospital were retrieved.2.Identify benign and malignant orbital neoplastic lesions by ultrasound diagnostic model and MRI,respectively.3.Analyze and compare the accuracy and consistency of the ultrasound diagnostic model and MRI in identifying benign and malignant orbital neoplastic lesions.Results :Part Ⅰ: Clinical and ultrasound characteristics analysis and diagnostic modeling of benign and malignant orbital neoplastic lesion1.The results of single-factor analysis showed that there were statistically significant differences in age,gender,number of involved quadrants,maximum diameter of tumor,morphology,aspect ratio,“cast” tumor,tumor “inverted triangle”,tumor “circular”,internal echo is very low,internal echo is uneven,internal strong echo content,and the proportion of blood flow richness(all P<0.05).2.Multi-factor logistic regression showed that strong internal echogenicity,very low solid partial echogenicity,“cast” tumor,age ≥56.5 years old,irregular morphology,male gender,and tumor involvement in ≥2 quadrants were independent predictors for malignant orbital tumors.The odds ratio(OR)was 5.715,4.767,4.665,3.628,3.597,2.383,and 1.928,respectively(all P < 0.05).A diagnostic model to identify benign and malignant orbital neoplastic lesions was established: logit(P)=-4.309 + 0.868 × male + 1.289 × age ≥ 56.5years + 0.657 × tumor involvement in ≥ 2 quadrants + 1.562 × solid partially echogenic very low + 1.280 × irregular morphology + 1.743 × strong internal echogenicity + 1.540 × tumor in a "cast" shape.3.ROC curve analysis showed that the area under the curve(area under the curve,AUC)of the diagnostic ultrasound model for differentiating benign and malignant orbital tumors was 0.925,with an accuracy of 86.1%,sensitivity of 69.3%,specificity of 92.0%,positive predictive value of 75.2%,negative predictive value of 89.5%,and Youden index of 0.613.Part Ⅱ: A comparative study on the value of ultrasonic diagnostic model and magnetic resonance imaging diagnosis of common orbital neoplastic lesions1.The accuracy of the ultrasound diagnostic model in identifying benign and malignant was 90.7%,and the accuracy of MRI in identifying benign and malignant was 91.0%,and the difference between them was not statistically significant(P > 0.05).This indicates that the diagnostic efficacy of the ultrasound diagnostic model in identifying benign and malignant orbital neoplastic lesions is comparable to that of MRI.2.The kappa concordance test between the ultrasound diagnostic model and MRI showed that the consistency between the ultrasound diagnostic model and MRI was good in identifying benign and malignant orbital neoplastic lesions(kappa=0.786,P<0.001).Conclusion:1.The diagnostic model based on ultrasound sonograms and clinical features is of value in identifying benign and malignant orbital neoplastic lesions.2.The ultrasonic diagnostic model and MRI have high diagnostic value in identifying benign and malignant orbital neoplastic lesions.
Keywords/Search Tags:Ultrasonography, Orbit, Neoplastic lesions, Magnetic resonance imaging, Benign and malignant
PDF Full Text Request
Related items