ABSTRACT
Objective To identify the eye movement characteristics that can distinguish reading interest and comprehensibilityin order to provide reference for prediction of reading interest and comprehensibility based on eye movement.Methods Tobii Glasses 3 was used to collect eye tracking data from 11 participants who read 12 texts on different topics.After reading all the texts,the participants marked each sentence with theirlevels of interest and comprehensibility.Then,Python R and excel software were used to preprocess and analyze the data,and the differences in eye movementamid the four combinations of reading interest(interested or not)and comprehensibility(levels of difficulty)were studiedat overall and individual levels.Results At the overall level,there were statistically significant differences in eye movement between the easy comprehension group and the difficult comprehensiongroup where interest was concerned,but there was little difference between the interested group and the disinterested groupwhen it came to comprehensibility.There were differences in eye movement behavior between individuals,and some of the statistically significant differences in eye movement were consistent across most of the participants.In addition,some of the eye movement characteristicswere oppositeacross different individuals or could not distinguish between the interested and disinterested at the overall level,but could be distinguished within individuals.Conclusion At the overall level,it is relatively easy to determine comprehensibility through eye movement,but it is more difficult to distinguish whether the participant is interested in the sentence or not,especially when the text is easily comprehensible.At the individual level,eye movement behavior varies widely between individuals,and individuals manifest different characteristics in distinguishing reading interest and comprehensibility.
ABSTRACT
Objective To analyze gene expression profiles of biopsy specimens from breast cancer patients who were treated with neoadjuvant chemotherapy(NAC) after biopsies, and to identify the genes which are closely associated with the efficacy of neoadjuvant chemotherapy with T/FAC [docetaxel(Taxotere), 5-fluorouracil, doxorubicin and cyclophosphamide] or T/FEC (Taxotere, 5-fluorouracil, epirubicin and cyclophosphamide) regimen.Methods We retrieved and collected gene expression profiles from publicly available databases.Four datasets, a total of 844 samples, were finally retained because all the patients had received a uniform neoadjuvant chemotherapy regimen.Response to neoadjuvant chemotherapy was categorized as a pathological complete response (pCR) or residual invasive cancer (RD).The differentially expressed genes (adjusted P-value<0.05) and therapeutic efficacy were analyzed and explored.Results After differential analysis, genes whose expressions were higher or lower in pCR group than in RD group were identified in each of the four datasets, respectively.There were 34 and 42 genes which were simultaneously more highly expressed or more lowly expressed in pCR group than in RD group in the four datasets.The unsupervised clustering, based on the 76 intersection genes, showed that the pCR specimens tended to form one cluster and the RD tended to form the other.Conclusion The seventy-six differentially expressed genes are associated with the efficacy of neoadjuvant chemotherapy and are likely to be novel predictive biomarkers for the efficacy of neoadjuvant chemotherapy.