ABSTRACT
BackgroundThe worldwide surge in coronavirus cases has led to the COVID-19 testing demand surge. Rapid, accurate, and cost-effective COVID-19 screening tests working at a population level are in imperative demand globally. MethodsBased on the eye symptoms of COVID-19, we developed and tested a COVID-19 rapid prescreening model using the eye-region images captured in China and Spain with cellphone cameras. The convolutional neural networks (CNNs)-based model was trained on these eye images to complete binary classification task of identifying the COVID-19 cases. The performance was measured using area under receiver-operating-characteristic curve (AUC), sensitivity, specificity, accuracy, and F1. The application programming interface was open access. FindingsThe multicenter study included 2436 pictures corresponding to 657 subjects (155 COVID-19 infection, 23{middle dot}6%) in development dataset (train and validation) and 2138 pictures corresponding to 478 subjects (64 COVID-19 infections, 13{middle dot}4%) in test dataset. The image-level performance of COVID-19 prescreening model in the China-Spain multicenter study achieved an AUC of 0{middle dot}913 (95% CI, 0{middle dot}898-0{middle dot}927), with a sensitivity of 0{middle dot}695 (95% CI, 0{middle dot}643-0{middle dot}748), a specificity of 0{middle dot}904 (95% CI, 0{middle dot}891-0{middle dot}919), an accuracy of 0{middle dot}875(0{middle dot}861-0{middle dot}889), and a F1 of 0{middle dot}611(0{middle dot}568-0{middle dot}655). InterpretationThe CNN-based model for COVID-19 rapid prescreening has reliable specificity and sensitivity. This system provides a low-cost, fully self-performed, non-invasive, real-time feedback solution for continuous surveillance and large-scale rapid prescreening for COVID-19. FundingThis project is supported by Aimomics (Shanghai) Intelligent
ABSTRACT
Objective To investigate the value of introvoxel incoherent motion(IVIM)using 3.0 T MRI to evaluate response to concurrent chemoradiotherapy(CCRT)in patients with advanced uterine cervix cancer. Methods From July 2015 to December 2016,63 patients with advanced(≥ⅡB)cervical cancer diagnosed by clinical and imaging study, who had completed CCRT plan in Henan Cancer Hospital, were prospectively enrolled.Pelvic MRI protocol including T1WI,T2WI,IVIM and dynamic contrasted enhanced scans were performed in each patient before CCRT and 3 weeks after starting therapy(total dose of 30 Gy), and at the end of therapy (total dose of 90 Gy, 8 weeks after therapy). The mean values of ADC, true molecular diffusion coefficient(D),pseudodiffusion coefficient(D*)and perfusion fraction(f)in each tumor at pre-therapy, in the middle of therapy and post-therapy were measured and recorded as ADC-pre, D-pre, D*-pre,f-pre;ADC-mid,D-mid,D*-mid,f-mid and ADC-post,D-post,D*-post,f-post,respectively;the change rates of these parameters during and after therapy (recorded as ΔADC-mid, ΔD-mid, ΔD*-mid, Δf-mid;ΔADC-post, ΔD-post, ΔD*-post, Δf-post) were also calculated. Patients were classified into response group and non-response group,according to response evaluation criteria in solid tumors after CCRT.MRI imaging study was performed in each patient within 1 month after CCRT to follow up,and tumor regression rate was calculated.The Mann-Whitney U test was used to compare differences of parameters and their change rates between response group and non-response group. Spearman correlation analysis was performed to assess relationships between parameters, parameter change rates and tumor regression rate. Logistic regression model was applied to find potential ADC values for predicting therapeutic response. ROC was used to analyze efficacy of ADC values for evaluating therapeutic response in advanced uterine cervix cancer after CCRT. Results The mean value of tumor maximum diameter before and after therapy was (47.5 ± 12.9) and(12.8 ± 10.0)mm,tumor regression rate was(66.7 ± 33.6)%.Forty-eight patients were in the response group and 15 in the non-response group.The mean value of ADC-pre,D-pre,D*-pre and f-pre was 0.74(0.43, 1.14)×10-3,0.58(0.33,0.91)×10-3,12.12(2.30,21.4)×10-3mm2/s,9.65%(4.45%,13.89%),respectively.Tumor regression rate had positive correlation with ADC-pre and D-pre (r=0.773,0.840;P<0.05). Responders had increased ADC-pre, D-pre values than non-responders, which had statistically significant difference (P<0.05). Responders had increased ADC-mid, D-mid and f-mid values than non-responders, which had statistically significant difference (P<0.05), tumor regression rate had positive correlation with ADC-mid, D-mid and f-mid (r=0.808,0.834,0.563;P<0.05). Responders had increased ADC-post, D-post and f-post values than non-responders,which had statistically significant difference(P<0.05),tumor regression rate had positive correlation with ADC-post and D-post (r=0.799, 0.829;P<0.05).Tumor regression rate had positive correlation with ΔADC-mid,ΔD-mid,Δf-mid(r=0.526,0.573,0.454;P<0.05)and with ΔADC-post,ΔD-post, Δf-post (r=0.541, 0.555, 0.388;P<0.05). Responders had increased ΔADC-mid, ΔD-mid, Δf-mid and ΔADC-post, ΔD-post, Δf-post, which had statistically significant difference (P<0.05). Logistic regression analysis revealed only ADC-pre and D-post could be independent factors to predict therapeutic response in advanced uterine cervix cancer after CCRT,values of B,Wald,odds ratio and P was 22.488,8.431,1.429, 0.004 and 16.542,8.517,1.779,0.004.ROC analysis showed the area under the curve(AUC)of ADC-pre, D-pre,ΔADC-mid,ΔD-mid,Δf-mid,ΔADC-post,ΔD-post and Δf-post for predicting therapeutic response in advanced uterine cervix cancer after CCRT were 0.890,0.926,0.942,0.851,0.803,0.929,0.951 and 0.906, respectively. Conclusion The IVIM parameters before and during CCRT process and their changes are valuable for predicting and evaluating therapeutic response in advanced uterine cervix cancer after CCRT, with high clinical practice value.
ABSTRACT
Objective To investigate the effect of respiratory motion on inadvertent irradiation dose (ⅡD)to the microscopic disease(MD)and expanding margin of target volume in stereotactic body radiotherapy for lung cancer. Methods Based on the pattern of respiration-induced tumor motion during lung radiotherapy, a probability model of MD entry into or exit from internal target volume(ITV)was established and the theoretical dose to MD was calculated according to the static dose distribution by four-dimensional computed tomography. The experimental dose to MD during respiratory motion was measured using a respiration simulation phantom and optically stimulated luminescence(OSL)and then compared with the theoretical value for model validation.Results For the target volume in periodic motion,the deviation of the theoretical dose to MD from the experimental value measured by OSL was less than 5%. A 10-mm margin around ITV received a biological dose higher than 80 Gy. Conclusions The dose model established in this study can accurately predict the irradiation dose to MD in the target volume in periodic motion. Respiratory motion increases ⅡD to MD and there is no need to expand clinical target volume.
ABSTRACT
As the neuron-specific actin binding protein,Drebrin ( developmentally regulated brain protein) can affect spiny and synaptic morphology and function by changing the property of cytoskeleton,and modulate synaptic plasticity. Neural excitability regulates expression and activitiy of Drebrin through a variety of signaling molecules,making Drebrin' s function correlated with that of neurons. Under pathological conditions,the abnormal Drebrin affects synaptic plasticity,which leads to different degrees of cognitive dysfunction,and it is closely relates to the progress of cognitive dysfunction. Extensive studies of physiological and pathological functions of Drebrin in overall and molecular ways will not only contribute to a thorough understanding of cognitive dysfunction,but also develop the new target of therapeutics for cognitive dysfunction.