Medical studies have found that tumor mutation burden (TMB) is positively correlated with the efficacy of immunotherapy for non-small cell lung cancer (NSCLC), and TMB value can be used to predict the efficacy of targeted therapy and chemotherapy. However, the calculation of TMB value mainly depends on the whole exon sequencing (WES) technology, which usually costs too much time and expenses. To deal with above problem, this paper studies the correlation between TMB and slice images by taking advantage of digital pathological slices commonly used in clinic and then predicts the patient TMB level accordingly. This paper proposes a deep learning model (RCA-MSAG) based on residual coordinate attention (RCA) structure and combined with multi-scale attention guidance (MSAG) module. The model takes ResNet-50 as the basic model and integrates coordinate attention (CA) into bottleneck module to capture the direction-aware and position-sensitive information, which makes the model able to locate and identify the interesting positions more accurately. And then, MSAG module is embedded into the network, which makes the model able to extract the deep features of lung cancer pathological sections and the interactive information between channels. The cancer genome map (TCGA) open dataset is adopted in the experiment, which consists of 200 pathological sections of lung adenocarcinoma, including 80 data samples with high TMB value, 77 data samples with medium TMB value and 43 data samples with low TMB value. Experimental results demonstrate that the accuracy, precision, recall and F1 score of the proposed model are 96.2%, 96.4%, 96.2% and 96.3%, respectively, which are superior to the existing mainstream deep learning models. The model proposed in this paper can promote clinical auxiliary diagnosis and has certain theoretical guiding significance for TMB prediction.
Objective To explore the risk factors of female’s breast cancer in secondary cities of the west and establish a risk prediction model to identify high-risk groups， and provide the basis for the primary and secondary preve-ntion of breast cancer. Methods Random sampling （method of random digits table） 1 700 women in secondary cities of the west （including 1 020 outpatient cases and 680 physical examination cases） were routinely accept the questionnaire survey. Sixty-two patients were confirmed breast cancer with pathologically. Based on the X-image of the mammary gland patients and questionnaire survey to put mammographic density which classificated into high- and low-density groups. The relationships between the mammographic density， age， body mass index （BMI）， family history of breast cancer， socio-economic status （SES）， lifestyle， reproductive fertility situation， and breast cancer were analyzed， then a risk prediction model of breast cancer which fitting related risk factors was established. Results Univariate analysis showed that risk factors for breast cancer were age （P=0.006）， BMI （P=0.007）， age at menarche （P=0.039）， occupation （P=0.001）， domicile place （P=0.000）， educational level （P=0.001）， health status compared to the previous year （P=0.046）， age at first birth （P=0.014）， whether menopause （P=0.003）， and age at menopause （P=0.006）. The unconditional logistic regr-ession analysis showed that the significant risk factors were age （P=0.003）， age at first birth （P=0.000）， occupation （P=0.010）， and domicile place （P=0.000）， and the protective factor was age at menarche （P=0.000）. The initially established risk prediction model in the region which fitting related risk factors was y=-5.557＋0.042x1-0.375x2＋1.206x3＋0.509x4＋2.135x5. The fitting coefficient （R square）=0.170， it could reflect 17% of the actual situation. Conclusions The breast cancer risk prediction model which established by using related risk factors analysis and epidemiological investigation could guide the future clinical work，but there is still need the validation studies of large populations for the model.