An integrated machine learning-based model for joint diagnosis of ovarian cancer with multiple test indicators.

OBJECTIVE: To construct a machine learning diagnostic model integrating feature dimensionality reduction techniques and artificial neural network classifiers to develop the value of clinical routine blood indexes for the auxiliary diagnosis of ovarian cancer. METHODS: Patients with ovarian cancer clearly diagnosed in our hospital were collected as a case group (n = 185), and three groups of patients with other malignant otolaryngology tumors (n = 138), patients with benign otolaryngology diseases (n = 339) and those with normal physical examination (n = 92) were used as an overall control group. In this paper, a fully automated segmentation network for magnetic resonance images of ovarian cancer is proposed to improve the reproducibility of tumor segmentation results while effectively reducing the burden on radiologists. A pre-trained Res Net50 is used to the three edge output modules are fused to obtain the final segmentation results. The segmentation results of the proposed network architecture are compared with the segmentation results of the U-net based network architecture and the effect of different loss functions and region of interest sizes on the segmentation performance of the proposed network is analyzed. RESULTS: The average Dice similarity coefficient, average sensitivity, average specificity (specificity) and average hausdorff distance of the proposed network segmentation results reached 83.62%, 89.11%, 96.37% and 8.50, respectively, which were better than the U-net based segmentation method. For ROIs containing tumor tissue, the smaller the size, the better the segmentation effect. Several loss functions do not differ much. The area under the ROC curve of the machine learning diagnostic model reached 0.948, with a sensitivity of 91.9% and a specificity of 86.9%, and its diagnostic efficacy was significantly better than that of the traditional way of detecting CA125 alone. The model was able to accurately diagnose ovarian cancer of different disease stages and showed certain discriminative ability for ovarian cancer in all three control subgroups. CONCLUSION: Using machine learning to integrate multiple conventional test indicators can effectively improve the diagnostic efficacy of ovarian cancer, which provides a new idea for the intelligent auxiliary diagnosis of ovarian cancer.

Mechanistic modelling of relative biological effectiveness of carbon ion beams and comparison with experiments.

Objective.Relative biological effectiveness (RBE) plays a vital role in carbon ion radiotherapy, which is a promising treatment method for reducing toxic effects on normal tissues and improving treatment efficacy. It is important to have an effective and precise way of obtaining RBE values to support clinical decisions. A method of calculating RBE from a mechanistic perspective is reported.Approach.Ratio of dose to obtain the same number of double strand breaks (DSBs) between different radiation types was used to evaluate RBE. Package gMicroMC was used to simulate DSB yields. The DSB inductions were then analyzed to calculate RBE. The RBE values were compared with experimental results.Main results.Furusawa's experiment yielded RBE values of 1.27, 2.22, 3.00 and 3.37 for carbon ion beam with dose-averaged LET of 30.3 keVµm-1, 54.5 keVµm-1, 88 keVµm-1and 137 keVµm-1, respectively. RBE values computed from gMicroMC simulations were 1.75, 2.22, 2.87 and 2.97. When it came to a more sophisticated carbon ion beam with 6 cm spread-out Bragg peak, RBE values were 1.61, 1.63, 2.19 and 2.36 for proximal, middle, distal and distal end part, respectively. Values simulated by gMicroMC were 1.50, 1.87, 2.19 and 2.34. The simulated results were in reasonable agreement with the experimental data.Significance.As a mechanistic way for the evaluation of RBE for carbon ion radiotherapy by combining the macroscopic simulation of energy spectrum and microscopic simulation of DNA damages, this work provides a promising tool for RBE calculation supporting clinical applications such as treatment planning.