Can deep learning replace histopathological examinations in the differential diagnosis of cervical lymphadenopathy?

INTRODUCTION: We aimed to develop a diagnostic deep learning model using contrast-enhanced CT images and to investigate whether cervical lymphadenopathies can be diagnosed with these deep learning methods without radiologist interpretations and histopathological examinations. MATERIAL METHOD: A total of 400 patients who underwent surgery for lymphadenopathy in the neck between 2010 and 2022 were retrospectively analyzed. They were examined in four groups of 100 patients: the granulomatous diseases group, the lymphoma group, the squamous cell tumor group, and the reactive hyperplasia group. The diagnoses of the patients were confirmed histopathologically. Two CT images from all the patients in each group were used in the study. The CT images were classified using ResNet50, NASNetMobile, and DenseNet121 architecture input. RESULTS: The classification accuracies obtained with ResNet50, DenseNet121, and NASNetMobile were 92.5%, 90.62, and 87.5, respectively. CONCLUSION: Deep learning is a useful diagnostic tool in diagnosing cervical lymphadenopathy. In the near future, many diseases could be diagnosed with deep learning models without radiologist interpretations and invasive examinations such as histopathological examinations. However, further studies with much larger case series are needed to develop accurate deep-learning models.

Predictors of allocation to surgery in patients older than 50 years with partial-thickness rotator cuff tear.

BACKGROUND: The purpose of this study was to determine the predictive factors for allocation to surgery in patients older than 50 years with symptomatic chronic partial-thickness rotator cuff tear (PTRCT). METHODS: Patients older than 50 years with a confirmed diagnosis of unilateral isolated PTRCT were included in this retrospective study. In the minimum follow-up of 2 years, eventual allocation to surgical or nonsurgical treatment was determined individually. Patients who underwent surgery were defined as failed conservative management and allocation to surgery. Data pertaining to patients' demographics, functional comorbidity index values, duration of symptoms, and American Shoulder and Elbow Surgeons scores were collected from our medical records. Tear side and Ellman classification, subacromial spur, and acromiohumeral intervals were also noted. A regression analysis was performed to determine the major predictors of allocation to surgery. RESULTS: There were 202 patients with a mean age of 62 years in group I (no-surgery group) and 70 patients with a mean age of 57 years in group II (surgery group). The mean age and functional comorbidity index values were significantly higher in group I than in group II (P < .001 and P < .001, respectively). Bursal-sided tears were significantly more common in group II (P = .026). According to the findings of regression analysis, tear side and functional comorbidity index were the major predictors of allocation to surgery (P = .015 and P < .001, respectively). CONCLUSION: Our study results indicate that in patients older than 50 years with PTRCTs, those with fewer comorbidities and bursal-sided PTRCTs were significantly more likely to undergo surgery.