Thyroid Imaging Reporting and Data System Score Combined with the New Italian Classification for Thyroid Cytology Improves the Clinical Management of Indeterminate Nodules.
Salvatore UlisseDaniela BoscoFrancesco NardiAngela NescaEleonora D'ArmientoValeria GuglielminoCorrado De VitoSalvatore SorrentiDaniele PironiFrancesco TartagliaStefano ArcieriAntonio CataniaMassimo MontiAngelo FilippiniValeria AscoliPublished in: International journal of endocrinology (2017)
The new Italian cytological classification (2014) of thyroid nodules replaced the TIR3 category of the old classification (2007) with two subclasses, TIR3A and TIR3B, with the aim of reducing the rate of surgery for benign diseases. Moreover, thyroid imaging reporting and data system (TI-RADS) score appears to ameliorate the stratification of the malignancy risk. We evaluated whether the new Italian classification has improved diagnostic accuracy and whether its association with TI-RADS score could improve malignancy prediction. We retrospectively analyzed 70 nodules from 70 patients classified as TIR3 according to the old Italian classification who underwent surgery for histological diagnosis. Of these, 51 were available for cytological revision according to the new Italian cytological classification. Risk of malignancy was determined for TIR3A and TIR3B, TI-RADS score, and their combination. A different rate of malignancy (p = 0.0286) between TIR3A (13.04%) and TIR3B (44.44%) was observed. Also TI-RADS score is significantly (p = 0.003) associated with malignancy. By combining cytology and TI-RADS score, patients could be divided into three groups with low (8.3%), intermediate (21.4%), and high (80%) risk of malignancy. In conclusion, the new Italian cytological classification has an improved diagnostic accuracy. Interestingly, the combination of cytology and TI-RADS score offers a better stratification of the malignancy risk.
Keyphrases
- deep learning
- fine needle aspiration
- machine learning
- end stage renal disease
- ejection fraction
- minimally invasive
- ultrasound guided
- newly diagnosed
- chronic kidney disease
- high grade
- high resolution
- prognostic factors
- big data
- artificial intelligence
- coronary artery bypass
- electronic health record
- adverse drug
- coronary artery disease
- patient reported outcomes
- total hip arthroplasty
- drug induced