Extraction of tongue carcinoma using genetic algorithm-induced fuzzy clustering and artificial neural network from MR images

A novel hierarchical image segmentation approach has been developed for the extraction of tongue carcinoma from magnetic resonance (MR) images. First, a genetic algorithm (GA)-induced fuzzy clustering is used for initial segmentation of MR images of head and neck. Then these segmented masses are refined to reduce the false-positives using an artificial neural network (ANN)-based symmetry detection and image analysis procedure. The proposed technique is applied to clinical MR images of tongue carcinoma and quantitative evaluations are performed. Experimental results suggest that the proposed approach provides an effective method for tongue carcinoma extraction with high accuracy and minimal user-dependency.

A benchtop system to assess cortical neural interface micromechanics

A benchtop brain tissue-microelectrode insertion model system was developed to aid in improving the design of cortical neural interfaces. The model partially mimics the in vivo environment via the use of human cadaver brain specimens (nspecimen = 6), or agar gel exposed to physiologically relevant mechanical oscillations. 150 lpm diameter stainless-steel microelectrode wires (TS = 600 MPa) implanted 3.0 cm within fixed human primary auditory cortex (ntrial > 10) experienced 133 +/- 8 and 64 +/- 4 mN of peak and steady axial forces. When subjected to a 3 Hz, 3-mm vertical oscillation, dynamic force amplitudes (ntrial > 10) of 148 +/- 10 mN were measured. The model system allows the study and comparison of static and dynamic forces and their mechanical influences on proposed implanted microelectrode structures.

Prevalence and temporal patterns of persistence of symptoms in 165 rectal cancer patients receiving preoperative chemoradiation therapy

19580

Background: The symptom burden during preoperative chemoradiation therapy for rectal cancer (CRT) has not been objectively characterized previously. The severity and temporal patterns of patient self-reported symptoms during CRT were assessed using the MD Anderson Symptom Inventory (MDASI). Methods: 165 consecutive patients with T3/T4/N+ rectal cancer received 45–55 Gy in 25–30 fractions concurrently with capecitabine chemotherapy. No additional intervention beyond standard supportive care was triggered by the MDASI score severity. The mean symptom scores were grouped into three time points; weeks 0–1 (“baseline”), 2–3 (“early treatment”), and 4–6 (“late treatment”) of CRT. Lowess curves were used to demonstrate the severity and pattern of individual symptoms. Linear mixed models were used to assess changes in symptom severity. Repeated measures analyses compared the mean severity of the reported symptoms at the three time points. Results: The median number of data points per patient was 3.0 (range, 1–6), for a total of 533 data points. Lowess curves demonstrated increasing symptom severity during CRT, which was confirmed by the linear mixed models analysis for each symptom (p<0.05). Using repeated measures with the three time points, these symptoms, except fatigue, showed statistically significant differences in mean severity ( Table 1 ). High baseline fatigue levels blunted statistical power to detect differences during CRT. Despite normal hemoglobin levels, fatigue remained the most severe symptom. Placed in context, the mean symptom levels remained mild throughout CRT. Conclusions: CRT was associated with a high prevalence of and progressive increase in symptom burden during therapy although the functional impact was limited. All symptoms followed clinically recognized patterns. The linear mixed model, a more sensitive measure of these trends, is recommended for prospective studies of interventions for symptom control

[Table: see text]

No significant financial relationships to disclose.