Leaf-level carbon isotope discrimination and its relationship with yield components as a tool for cotton phenotyping in unfavorable conditions

Abstract

OBJECTIVE: The objective of this study was to investigate the feasibilityof computerized segmentation of lymph nodes to evaluate the relationshipof treatment outcomes in tuberculous cervical lymphadenitis (TCL).METHODS: The study included 52 subjects with TCL that had CT beforestandard medical treatment. The relationship between the medicationtreatment response, volume and ratio of necrotic area of the largestlymph node in patients with TCL was evaluated. The treatment outcomewas defined as the 'responder' (n=35) and 'non-responder' (n=17)groups. Seventeen non-responder patients required surgery. RESULTS:The average lymph node volumes and ratio of necrotic area were 58.59cm(3)(range, 4.96-249.48cm(3)) and 0.30 (range, 0-0.59), respectively.There was a significant difference in the lymph node volumes (34.91�24.00cm(3)vs. 107.04�69.12cm(3), p<0.001) and ratio of necrotic area (0.26�0.12vs. 0.40�0.14, p=0.001) between responders and non-responders. Thereceiver-operating-characteristic (ROC) curve analysis was used fordifferentiating responders from non-responders; it showed that thearea under the curve for the lymph node volumes and ratio of necroticarea was 0.845 and 0.759, respectively. The cut-off value for thelymph node volumes and ratio of necrotic area was 44.15cm(3) and0.36 based on the ROC curve. CONCLUSIONS: A large lymph node volumeand high ratio of necrotic area on the 3D reconstruction of CT imageswere associated with the response to medical treatment for TB. Thesefindings might be useful for assessing treatment outcomes.