Deep learning for heterogeneous medical data analysis

Yue, Lin; Tian, Di; Chen, Weitong; Han, Xuming; Yin, Minghao

doi:10.1007/s11280-019-00764-z

Cited by 54 publications

(15 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current methods primarily deal with finding markers of diseases, reporting and detecting human diseases, and monitoring elderly people [56]. For example, DL can involve the heterogenous data of elderly patients [57], use of wearable technology [51] to monitor individual health status, such as cardiac rhythms and movements and their daily lifestyle [58], biometric data [59], and human genes [48], etc., and can advance the analysis of such unstructured data to relevant diagnostic information by removing anomalies and deriving useful patterns.…”

Section: Methodsmentioning

confidence: 99%

Role of Deep Learning in Predicting Aging-Related Diseases: A Scoping Review

Wassan

Zheng

Wang

2021

Cells

View full text Add to dashboard Cite

Aging refers to progressive physiological changes in a cell, an organ, or the whole body of an individual, over time. Aging-related diseases are highly prevalent and could impact an individual’s physical health. Recently, artificial intelligence (AI) methods have been used to predict aging-related diseases and issues, aiding clinical providers in decision-making based on patient’s medical records. Deep learning (DL), as one of the most recent generations of AI technologies, has embraced rapid progress in the early prediction and classification of aging-related issues. In this paper, a scoping review of publications using DL approaches to predict common aging-related diseases (such as age-related macular degeneration, cardiovascular and respiratory diseases, arthritis, Alzheimer’s and lifestyle patterns related to disease progression), was performed. Google Scholar, IEEE and PubMed are used to search DL papers on common aging-related issues published between January 2017 and August 2021. These papers were reviewed, evaluated, and the findings were summarized. Overall, 34 studies met the inclusion criteria. These studies indicate that DL could help clinicians in diagnosing disease at its early stages by mapping diagnostic predictions into observable clinical presentations; and achieving high predictive performance (e.g., more than 90% accurate predictions of diseases in aging).

show abstract

Section: Methodsmentioning

confidence: 99%

Role of Deep Learning in Predicting Aging-Related Diseases: A Scoping Review

Wassan

Zheng

Wang

2021

Cells

View full text Add to dashboard Cite

show abstract

“…In the last years, CNNs have obtained several highprofile results in a variety of medical fields, ranging from classification of skin cancer in dermatology [11] to detecting diabetic retinopathy [10]. Further interest has been gained with the emergence of powerful methods for performing automatic segmentation of 2D [18] and 3D [19] biomedical data and for exploiting heterogeneous sources of information [20]. We refer the interested reader to specific surveys [12,[21][22][23][24] for an overview of CNN applications in medical histology and cytology.…”

Section: Deep Network For Medicalmentioning

confidence: 99%

A Calibrated Multiexit Neural Network for Detecting Urothelial Cancer Cells

Lilli

Giarnieri

Scardapane

2021

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

Deep convolutional networks have become a powerful tool for medical imaging diagnostic. In pathology, most efforts have been focused in the subfield of histology, while cytopathology (which studies diagnostic tools at the cellular level) remains underexplored. In this paper, we propose a novel deep learning model for cancer detection from urinary cytopathology screening images. We leverage recent ideas from the field of multioutput neural networks to provide a model that can efficiently train even on small-scale datasets, such as those typically found in real-world scenarios. Additionally, we argue that calibration (i.e., providing confidence levels that are aligned with the ground truth probability of an event) has been a major shortcoming of prior works, and we experiment a number of techniques to provide a well-calibrated model. We evaluate the proposed algorithm on a novel dataset, and we show that the combination of focal loss, multiple outputs, and temperature scaling provides a model that is significantly more accurate and calibrated than a baseline deep convolutional network.

show abstract

“…Zhang et al demonstrated that the models constructed based on the integration of unstructured clinical notes with structured data outperformed other models that utilize only unstructured notes or structured data [96]. Other DL methods have recently been reviewed for heterogeneous medical data [97] and image analysis [98].…”

Section: High-data Quality Are Required For Accurate Prediction Modelsmentioning

confidence: 99%

Personalized Cell Therapy for Patients With Peripheral Arterial Diseases in the Context of Genetic Alterations: Artificial Intelligence-based Responder and Non-responder Prediction

Salybekov¹,

Wolfien²,

Kobayashi³

et al. 2021

Preprint

View full text Add to dashboard Cite

Stem/progenitor cell transplantation is a potential novel therapeutic strategy to induce angiogenesis in ischemic tissue, which can prevent major amputation in patients with advanced peripheral artery disease (PAD). Thus, clinicians can use cell therapies worldwide to treat PAD. However, some cell therapy studies did not report beneficial outcomes. Clinical researchers suggested that classical risk factors and comorbidities may adversely affect the efficacy of cell therapy. Some studies have indicated that the response to stem cell therapy varies among patients even in those harboring limited risk factors. This suggested the role of undetermined risk factors, including genetic alterations, somatic mutations, and clonal hematopoiesis. Personalized stem cell-based therapy can be developed by analyzing individual risk factors. These approaches must consider several clinical biomarkers and perform studies (such as genome-wide association studies (GWAS)) on disease-related genetic traits and integrate the findings with those of transcriptome-wide association studies (TWAS) and whole-genome sequencing in PAD. Additional unbiased analyses with state-of-the-art computational methods, such as machine learning-based patient stratification, are suited for predictions in clinical investigations. The integration of these complex approaches into a unified analysis procedure for the identification of responders and non-responders before stem cell therapy, which can decrease treatment expenditure, is a major challenge to increase the efficacy of therapies.

show abstract

Deep learning for heterogeneous medical data analysis

Cited by 54 publications

References 61 publications

Role of Deep Learning in Predicting Aging-Related Diseases: A Scoping Review

Role of Deep Learning in Predicting Aging-Related Diseases: A Scoping Review

A Calibrated Multiexit Neural Network for Detecting Urothelial Cancer Cells

Personalized Cell Therapy for Patients With Peripheral Arterial Diseases in the Context of Genetic Alterations: Artificial Intelligence-based Responder and Non-responder Prediction

Contact Info

Product

Resources

About