Tridirectional Transfer Learning for Predicting Gastric Cancer Morbidity

A Transfer Learning Radiomics Nomogram to Predict the Postoperative Recurrence of Advanced Gastric Cancer

BACKGROUND AND AIM

In this study, a transfer learning (TL) algorithm was used to predict postoperative recurrence of advanced gastric cancer (AGC) and to evaluate its value in a small-sample clinical study.

METHODS

A total of 431 cases of AGC from three centers were included in this retrospective study. First, TL signatures (TLSs) were constructed based on different source domains, including whole slide images (TLS-WSIs) and natural images (TLS-ImageNet). Clinical model and non-TLS based on CT images were constructed simultaneously. Second, TL radiomic model (TLRM) was constructed by combining optimal TLS and clinical factors. Finally, the performance of the models was evaluated by ROC analysis. The clinical utility of the models was assessed using integrated discriminant improvement (IDI) and decision curve analysis (DCA).

RESULTS

TLS-WSI significantly outperformed TLS-ImageNet, non-TLS, and clinical models (p < 0.05). The AUC value of TLS-WSI in training cohort was 0.9459 (95CI%: 0.9054, 0.9863) and ranged from 0.8050 (95CI%: 0.7130, 0.8969) to 0.8984 (95CI%: 0.8420, 0.9547) in validation cohorts. TLS-WSI and the nodular or irregular outer layer of gastric wall were screened to construct TLRM. The AUC value of TLRM in training cohort was 0.9643 (95CI%: 0.9349, 0.9936) and ranged from 0.8561 (95CI%: 0.7571, 0.9552) to 0.9195 (95CI%: 0.8670, 0.9721) in validation cohorts. The IDI and DCA showed that the performance of TLRM outperformed the other models.

CONCLUSION

TLS-WSI can be used to predict postoperative recurrence in AGC, whereas TLRM is more effective. TL can effectively improve the performance of clinical research models with a small sample size.

Crop cultivation planning with fuzzy estimation using water wave optimization

In a complex agricultural region, determine the appropriate crop for each plot of land to maximize the expected total profit is the key problem in cultivation management. However, many factors such as cost, yield, and selling price are typically uncertain, which causes an exact programming method impractical. In this paper, we present a problem of crop cultivation planning, where the uncertain factors are estimated as fuzzy parameters. We adapt an efficient evolutionary algorithm, water wave optimization (WWO), to solve this problem, where each solution is evaluated based on three metrics including the expected, optimistic and pessimistic values, the combination of which enables the algorithm to search credible solutions under uncertain conditions. Test results on a set of agricultural regions in East China showed that the solutions of our fuzzy optimization approach obtained significantly higher profits than those of non-fuzzy optimization methods based on only the expected values.

Classification of tumor from computed tomography images: A brain-inspired multisource transfer learning under probability distribution adaptation

Preoperative diagnosis of gastric cancer and primary gastric lymphoma is challenging and has important clinical significance. Inspired by the inductive reasoning learning of the human brain, transfer learning can improve diagnosis performance of target task by utilizing the knowledge learned from the other domains (source domain). However, most studies focus on single-source transfer learning and may lead to model performance degradation when a large domain shift exists between the single-source domain and target domain. By simulating the multi-modal information learning and transfer mechanism of human brain, this study designed a multisource transfer learning feature extraction and classification framework, which can enhance the prediction performance of the target model by using multisource medical data (domain). First, this manuscript designs a feature extraction network that takes the maximum mean difference based on the Wasserstein distance as an adaptive measure of probability distribution and extracts the domain-specific invariant representations between source and target domain data. Then, aiming at the random generation of parameters bringing uncertainties to prediction accuracy and generalization ability of extreme learning machine network, the 1-norm regularization is used to implement sparse constraints of the output weight matrix and improve the robustness of the model. Finally, some experiments are carried out on the data of two medical centers. The experimental results show that the area under curves (AUCs) of the method are 0.958 and 0.929 in the two validation cohorts, respectively. The method in this manuscript can provide doctors with a better diagnostic reference, which has certain practical significance.

Design and Development of Modified Ensemble Learning with Weighted RBM Features for Enhanced Multi-disease Prediction Model

In this computer world, huge data are generated in several fields. Statistics in the healthcare engineering provides data about many diseases and corresponding patient's information. These data help to evaluate a huge amount of data for identifying the unknown patterns in the diseases and are also utilized for predicting the disease. Hence, this work is to plan and implement a new computer-aided technique named modified Ensemble Learning with Weighted RBM Features (EL-WRBM). Data collection is an initial process, in which the data of various diseases are gathered from UCI repository and Kaggle. Then, the gathered data are pre-processed by missing data filling technique. Then, the pre-processed data are performed by deep belief network (DBN), in which the weighted features are extracted from the RBM regions. Then, the prediction is made by ensemble learning with classifiers, namely, support vector machine (SVM), recurrent neural network (RNN), and deep neural network (DNN), in which hyper-parameters are optimized by the adaptive spreading rate-based coronavirus herd immunity optimizer (ASR-CHIO). At the end, the simulation analysis reveals that the suggested model has implications to support doctor diagnoses.

Cyborg Moth Flight Control Based on Fuzzy Deep Learning

Cyborg insect control methods can be divided into invasive methods and noninvasive methods. Compared to invasive methods, noninvasive methods are much easier to implement, but they are sensitive to complex and highly uncertain environments, for which classical control methods often have low control accuracy. In this paper, we present a noninvasive approach for cyborg moths stimulated by noninvasive ultraviolet (UV) rays. We propose a fuzzy deep learning method for cyborg moth flight control, which consists of a Behavior Learner and a Control Learner. The Behavior Learner is further divided into three hierarchies for learning the species’ common behaviors, group-specific behaviors, and individual-specific behaviors step by step to produce the expected flight parameters. The Control Learner learns how to set UV ray stimulation to make a moth exhibit the expected flight behaviors. Both the Control Learner and Behavior Learner (including its sub-learners) are constructed using a Pythagorean fuzzy denoising autoencoder model. Experimental results demonstrate that the proposed approach achieves significant performance advantages over the state-of-the-art approaches and obtains a high control success rate of over 83% for flight parameter control.

Transfer learning for non-image data in clinical research: A scoping review

BACKGROUND

Transfer learning is a form of machine learning where a pre-trained model trained on a specific task is reused as a starting point and tailored to another task in a different dataset. While transfer learning has garnered considerable attention in medical image analysis, its use for clinical non-image data is not well studied. Therefore, the objective of this scoping review was to explore the use of transfer learning for non-image data in the clinical literature.

METHODS AND FINDINGS

We systematically searched medical databases (PubMed, EMBASE, CINAHL) for peer-reviewed clinical studies that used transfer learning on human non-image data. We included 83 studies in the review. More than half of the studies (63%) were published within 12 months of the search. Transfer learning was most often applied to time series data (61%), followed by tabular data (18%), audio (12%) and text (8%). Thirty-three (40%) studies applied an image-based model to non-image data after transforming data into images (e.g. spectrograms). Twenty-nine (35%) studies did not have any authors with a health-related affiliation. Many studies used publicly available datasets (66%) and models (49%), but fewer shared their code (27%).

CONCLUSIONS

In this scoping review, we have described current trends in the use of transfer learning for non-image data in the clinical literature. We found that the use of transfer learning has grown rapidly within the last few years. We have identified studies and demonstrated the potential of transfer learning in clinical research in a wide range of medical specialties. More interdisciplinary collaborations and the wider adaption of reproducible research principles are needed to increase the impact of transfer learning in clinical research.

Designing deep learning studies in cancer diagnostics

The number of publications on deep learning for cancer diagnostics is rapidly increasing, and systems are frequently claimed to perform comparable with or better than clinicians. However, few systems have yet demonstrated real-world medical utility. In this Perspective, we discuss reasons for the moderate progress and describe remedies designed to facilitate transition to the clinic. Recent, presumably influential, deep learning studies in cancer diagnostics, of which the vast majority used images as input to the system, are evaluated to reveal the status of the field. By manipulating real data, we then exemplify that much and varied training data facilitate the generalizability of neural networks and thus the ability to use them clinically. To reduce the risk of biased performance estimation of deep learning systems, we advocate evaluation in external cohorts and strongly advise that the planned analyses, including a predefined primary analysis, are described in a protocol preferentially stored in an online repository. Recommended protocol items should be established for the field, and we present our suggestions.

Multi-Objective Optimization of Integrated Civilian-Military Scheduling of Medical Supplies for Epidemic Prevention and Control

In a large-scale epidemic, such as the novel coronavirus pneumonia (COVID-19), there is huge demand for a variety of medical supplies, such as medical masks, ventilators, and sickbeds. Resources from civilian medical services are often not sufficient for fully satisfying all of these demands. Resources from military medical services, which are normally reserved for military use, can be an effective supplement to these demands. In this paper, we formulate a problem of integrated civilian-military scheduling of medical supplies for epidemic prevention and control, the aim of which is to simultaneously maximize the overall satisfaction rate of the medical supplies and minimize the total scheduling cost, while keeping a minimum ratio of medical supplies reservation for military use. We propose a multi-objective water wave optimization (WWO) algorithm in order to efficiently solve this problem. Computational results on a set of problem instances constructed based on real COVID-19 data demonstrate the effectiveness of the proposed method.