Risk adjustment for measuring health care outcomes

Abstract Carcinoma is a primary source of morbidity in women globally, with metastatic disease accounting for most deaths. Its early discovery and diagnosis may significantly increase the odds of survival. Breast cancer imaging is critical for early identification, clinical staging, management choices, and treatment planning. In the current study, the FastAI technology is used with the ResNet-32 model to precisely identify ductal carcinoma. ResNet-32 is having few layers comparted to majority of its counterparts with almost identical performance. FastAI offers a rapid approximation toward the outcome for deep learning models via GPU acceleration and a faster callback mechanism, which would result in faster execution of the model with lesser code and yield better precision in classifying the tissue slides. Residual Network (ResNet) is proven to handle the vanishing gradient and effective feature learning better. Integration of two computationally efficient technologies has yielded a precision accuracy with reasonable computational efforts. The proposed model has shown considerable efficiency in the evaluating parameters like sensitivity, specificity, accuracy, and F1 Score against the other dominantly used deep learning models. These insights have shown that the proposed approach might assist practitioners in analyzing Breast Cancer (BC) cases appropriately, perhaps saving future complications and death. Clinical and pathological analysis and predictive accuracy have been improved with digital image processing. 

/pdf/resnet-32-and-fastai-for-diagnoses-of-ductal-carcinoma-from-31i3a05q.pdf

ResNet-32 and FastAI for diagnoses of ductal carcinoma from 2D tissue slides

The main task of medical image registration is to match different modal images or the same modal images of different periods to provide the doctor with richer diagnostic information. Image registration has been widely used in image diagnostics, image-guided surgical planning and real-time interventional surgical navigation. The deformation model is a key part of the image registration process and can drive the image deformation to achieve a perfect match of the same organization in the two images. In practical application, it is important to establish a reasonable registration deformation model according to the research object, which directly affects the registration results. This paper presents a review of the deformation models used in medical image registration. The deformation model is summarized with respect to four aspects: the elastic image model, the viscoelastic image model, the optical flow model and the prior knowledge model. We primarily summarize the deformation models with good registration results in recent years and analyze their adaptability and existing defects. The purpose of this paper is to provide a reference for the selection of a deformation model.

A Review of Deformation Models in Medical Image Registration

Estimation of fiber orientations using neighborhood information.

In this paper, we address two research topics in Recommender Systems (RSs) which have been developed in parallel without a deeper integration: Cross-Domain RS (CDRS) and Context-Aware RS (CARS). CDRS have emerged to enhance the quality of recommendations in a target domain by leveraging sources of information in different domains. CDRS are especially useful to address cold-start, sparsity and diversity problems in target domains with scarce information. CARS, on its turn, have been proposed to consider contextual information for recommendations. Such systems are suitable when the users’ interests change according to factors like time, location, among others. By combining these two approaches, better RSs can be developed, considering both the availability of useful data from multiple domains and the use of contextual information. In this paper, we formalize the combination of CDRS and CARS, which represents a more systematic integration of these approaches compared to previous work. Based on this formulation, we developed novel RSs techniques, named CD-CARS. To evaluate the developed CD-CARS techniques, we performed extensive experimentation through real datasets taking into account several scenarios. The recommendations were evaluated in terms of predictive and ranking performance, respectively achieving up to 62.6% and 45%, depending on the scenario, in comparison to traditional cross-domain collaborative filtering techniques. Therefore, the experimental results have shown that the integration of techniques developed in isolation can be useful in a variety of situations, in which recommendations can be improved by information gathered from different sources and can be refined by considering specific contextual information.

https://repositorio.ufpe.br/bitstream/123456789/18356/1/dvsTeseBiblioteca.pdf

CD-CARS: Cross-domain context-aware recommender systems

Breast cancer has high incidences and mortality rates in women worldwide. Malignancy could be detected manually by experienced pathologists based on Hematoxylin and Eosin (H&E) stained images. However, it is time‐consuming and experience‐dependent, making early diagnosis a big challenge. In this paper, a methodology for breast cancer classification based on histopathological images with deep learning was described. A residual learning‐based convolutional neural network named myResNet‐34 was designed for malignancy‐and‐benign classification. In addition, an algorithm automatically generating the target image for stain normalization was proposed, which eliminated the bias caused by manual selection of the reference image. Elastic distortion was introduced and combined with affine transformation for data augmentation considering the characteristics of the H&E images. Experiments were conducted on BreakHis dataset with the proposed framework. Promising results were achieved with an average classification accuracy of around 91% on image‐level classification. Results indicated that both our data augmentation and stain normalization effectively improved the classification accuracy by 2‐3%.

Classification of breast cancer histopathological image with deep residual learning

Alzheimer's disease (AD) is a neurodegenerative disease that is difficult to be detected using convenient and reliable methods. The language change in patients with AD is an important signal of their cognitive status, which potentially helps in early diagnosis. In this study, we developed a transfer learning model based on speech and natural language processing (NLP) technology for the early diagnosis of AD. The lack of large datasets limits the use of complex neural network models without feature engineering, while transfer learning can effectively solve this problem. The transfer learning model is firstly pre-trained on large text datasets to get the pre-trained language model, and then, based on such a model, an AD classification model is performed on small training sets. Concretely, a distilled bidirectional encoder representation (distilBert) embedding, combined with a logistic regression classifier, is used to distinguish AD from normal controls. The model experiment was evaluated on Alzheimer's dementia recognition through spontaneous speech datasets in 2020, including the balanced 78 healthy controls (HC) and 78 patients with AD. The accuracy of the proposed model is 0.88, which is almost equivalent to the champion score in the challenge and a considerable improvement over the baseline of 75% established by organizers of the challenge. As a result, the transfer learning method in this study improves AD prediction, which does not only reduces the need for feature engineering but also addresses the lack of sufficiently large datasets.

/pdf/a-transfer-learning-method-for-detecting-alzheimer-s-disease-2z448m9d.pdf

A Transfer Learning Method for Detecting Alzheimer's Disease Based on Speech and Natural Language Processing

The demons algorithm is a popular algorithm for non-rigid image registration because of its computational efficiency and simple implementation. The deformation forces of the classic demons algorithm were derived from image gradients by considering the deformation to decrease the intensity dissimilarity between images. However, the methods using the difference of image intensity for medical image registration are easily affected by image artifacts, such as image noise, non-uniform imaging and partial volume effects. The gradient magnitude image is constructed from the local information of an image, so the difference in a gradient magnitude image can be regarded as more reliable and robust for these artifacts. Then, registering medical images by considering the differences in both image intensity and gradient magnitude is a straightforward selection. In this paper, based on a diffeomorphic demons algorithm, we propose a chain-type diffeomorphic demons algorithm by combining the differences in both image intensity and gradient magnitude for medical image registration. Previous work had shown that the classic demons algorithm can be considered as an approximation of a second order gradient descent on the sum of the squared intensity differences. By optimizing the new dissimilarity criteria, we also present a set of new demons forces which were derived from the gradients of the image and gradient magnitude image. We show that, in controlled experiments, this advantage is confirmed, and yields a fast convergence.

https://iopscience.iop.org/article/10.1088/0031-9155/58/23/8359/pdf

Medical image registration by combining global and local information: a chain-type diffeomorphic demons algorithm

Purpose: Diffusion tensor imaging is widely used for studying neural fiber trajectories in white matter and for quantifying changes in tissue using diffusion properties at each voxel in the brain. To better model the nature of crossing fibers within complex architectures, rather than using a simplified tensor model that assumes only a single fiber direction at each image voxel, a model mixing multiple diffusion tensors is used to profile diffusion signals from high angular resolution diffusion imaging (HARDI) data. Based on the HARDI signal and a multiple tensors model, spherical deconvolution methods have been developed to overcome the limitations of the diffusion tensor model when resolving crossing fibers. The Richardson–Lucy algorithm is a popular spherical deconvolution method used in previous work. However, it is based on a Gaussian distribution, while HARDI data are always very noisy, and the distribution of HARDI data follows a Rician distribution. This current work aims to present a novel solution to address these issues. Methods: By simultaneously considering both the Rician bias and neighbor correlation in HARDI data, the authors propose a localized Richardson–Lucy (LRL) algorithm to estimate fiber orientations for HARDI data. The proposed method can simultaneously reduce noise and correct the Rician bias. Results: Mean angular error (MAE) between the estimated Fiber orientation distribution (FOD) field and the reference FOD field was computed to examine whether the proposed LRL algorithm offered any advantage over the conventional RL algorithm at various levels of noise. Normalized mean squared error (NMSE) was also computed to measure the similarity between the true FOD field and the estimated FOD filed. For MAE comparisons, the proposed LRL approach obtained the best results in most of the cases at different levels of SNR and b-values. For NMSE comparisons, the proposed LRL approach obtained the best results in most of the cases at b-value = 3000 s/mm2, which is the recommended schema for HARDI data acquisition. In addition, the FOD fields estimated by the proposed LRL approach in regions of fiber crossing regions using real data sets also showed similar fiber structures which agreed with common acknowledge in these regions. Conclusions: The novel spherical deconvolution method for improved accuracy in investigating crossing fibers can simultaneously reduce noise and correct Rician bias. With the noise smoothed and bias corrected, this algorithm is especially suitable for estimation of fiber orientations in HARDI data. Experimental results using both synthetic and real imaging data demonstrated the success and effectiveness of the proposed LRL algorithm.

A localized Richardson-Lucy algorithm for fiber orientation estimation in high angular resolution diffusion imaging.

Rapidly increasing volumes of heterogeneous media digital contents are produced into the digital library by the forms of the digital books, videos, images, etc. However, traditional recommendation approaches in the digital library cannot support the potential semantic connections across different types of media data. In this paper, a cross-media recommendation algorithm for the digital library is proposed, in which the retrieved items may come from different data sources, and the results do not need to be of the same media type the user ever read or tagged. Firstly, a fused user-item-feature tensor is used to represent the cross-media data set. Then the item-context latent space and item-user rating latent space are reconstructed by TUCKER based tensor decomposition. And the structural grouping sparsity approach is used to select the feature groups and the subset of homogeneous features in one group, which can deal with the difficulty of sparse and high dimension of the big feature matrix. Finally, the Top-n items are recommended according to the prediction probability estimated. Experiments conducted on a cross-media dataset based on China Academic Digital Associative Library (CADAL). The performances evaluation is based on the recall precision and diversity score. The experiment results show that our approach has good recommendation accuracy as well as good diversity.

Zhenming Yuan

Papers

Classification of breast cancer histopathological image with deep residual learning

A Transfer Learning Method for Detecting Alzheimer's Disease Based on Speech and Natural Language Processing

Medical image registration by combining global and local information: a chain-type diffeomorphic demons algorithm

A localized Richardson-Lucy algorithm for fiber orientation estimation in high angular resolution diffusion imaging.

Cross Media Recommendation in Digital Library