Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis.
Ontology highlight
ABSTRACT:
SUBMITTER: Du-Harpur X
PROVIDER: S-EPMC7990050 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Background and aimsMicroRNAs (miRNAs) have been shown to play important roles in many cancers, including breast cancer. The majority of previous studies employed network analysis to identify key miRNAs in cancer progression. However, most of dysregulated miRNA networks were constructed based on the expression variation of miRNAs and target genes.MethodsThe relations between miRNAs and target genes were computed by Spearman correlation separately in breast cancer and normal breast samples. We calculated dysregulated scores based on the dysregulation of miRNA-mRNA regulatory relations. A dysregulated miRNA target network (DMTN) was constructed from the miRNA-mRNA pairs with significant dysregulated scores. SVM classifier was employed to predict breast cancer risk miRNAs from the DMTN. Hypermetric test was utilized to calculate the significance of overlap between different gene sets. Pearson correlation was used to evaluate associations between miRNAs/genes and drug response.ResultsThe DMTN comprised 511 miRNAs and was similar to common biological networks. Based on miRNAs and target genes in DMTN, we predicted 90 breast cancer risk miRNAs by using SVM classifier. Predicted risk miRNAs and one-step neighbor genes were significantly overlapping with differential miRNAs, cancer-related and housekeeping genes in breast cancer. These risk miRNAs were involved in many cancer-related and immune-related processes. In addition, most risk miRNAs were able to predict survival of breast cancer patients. More interestingly, some risk miRNAs and one-step neighbor genes were remarkably associated with immune cell infiltration. For example, high expression of hsa-miR-155 indicates high abundance of activated CD4+ T cells but low level of M2 macrophage infiltration. Furthermore, we identified 588 miRNA-drug and 3,146 gene-drug pairs, wherein the expression level of miRNAs/genes could indicate the sensitivity of cancer cells to anti-cancer drugs.ConclusionWe predicted 90 breast cancer risk miRNAs based on proposed DMTN by using SVM classifier. Predicted risk miRNAs are biologically and clinically relevant in breast cancer. Risk miRNAs and one-step neighbor genes could serve as biomarkers for immune cell infiltration and anti-cancer drug response, which sheds lights on immunotherapy or targeted therapy for patients with breast cancer.
| S-EPMC7689188 | biostudies-literature
Project description:The early detection and accurate histopathological diagnosis of gastric cancer increase the chances of successful treatment. The worldwide shortage of pathologists offers a unique opportunity for the use of artificial intelligence assistance systems to alleviate the workload and increase diagnostic accuracy. Here, we report a clinically applicable system developed at the Chinese PLA General Hospital, China, using a deep convolutional neural network trained with 2,123 pixel-level annotated H&E-stained whole slide images. The model achieves a sensitivity near 100% and an average specificity of 80.6% on a real-world test dataset with 3,212 whole slide images digitalized by three scanners. We show that the system could aid pathologists in improving diagnostic accuracy and preventing misdiagnoses. Moreover, we demonstrate that our system performs robustly with 1,582 whole slide images from two other medical centres. Our study suggests the feasibility and benefits of using histopathological artificial intelligence assistance systems in routine practice scenarios.
| S-EPMC7453200 | biostudies-literature
Project description:Skin diseases constitute a widespread health concern, and the application of machine learning and deep learning algorithms has been instrumental in improving diagnostic accuracy and treatment effectiveness. This paper aims to provide a comprehensive review of the existing research on the utilization of machine learning and deep learning in the field of skin disease diagnosis, with a particular focus on recent widely used methods of deep learning. The present challenges and constraints were also analyzed and possible solutions were proposed. We collected comprehensive works from the literature, sourced from distinguished databases including IEEE, Springer, Web of Science, and PubMed, with a particular emphasis on the most recent 5-year advancements. From the extensive corpus of available research, twenty-nine articles relevant to the segmentation of dermatological images and forty-five articles about the classification of dermatological images were incorporated into this review. These articles were systematically categorized into two classes based on the computational algorithms utilized: traditional machine learning algorithms and deep learning algorithms. An in-depth comparative analysis was carried out, based on the employed methodologies and their corresponding outcomes. Present outcomes of research highlight the enhanced effectiveness of deep learning methods over traditional machine learning techniques in the field of dermatological diagnosis. Nevertheless, there remains significant scope for improvement, especially in improving the accuracy of algorithms. The challenges associated with the availability of diverse datasets, the generalizability of segmentation and classification models, and the interpretability of models also continue to be pressing issues. Moreover, the focus of future research should be appropriately shifted. A significant amount of existing research is primarily focused on melanoma, and consequently there is a need to broaden the field of pigmented dermatology research in the future. These insights not only emphasize the potential of deep learning in dermatological diagnosis but also highlight directions that should be focused on.
| S-EPMC10706240 | biostudies-literature
Project description:The molecular events and transcriptional plasticity driving brain metastasis in clinically relevant breast tumor subtypes has not been determined. Here we comprehensively dissect genomic, transcriptomic and clinical data in patient-matched longitudinal tumor samples, and unravel distinct transcriptional programs enriched in brain metastasis. We report on subtype specific hub genes and functional processes, central to disease-affected networks in brain metastasis. Importantly, in luminal brain metastases we identify homologous recombination deficiency operative in transcriptomic and genomic data with recurrent breast mutational signatures A, F and K, associated with mismatch repair defects, TP53 mutations and homologous recombination deficiency (HRD) respectively. Utilizing PARP inhibition in patient-derived brain metastatic tumor explants we functionally validate HRD as a key vulnerability. Here, we demonstrate a functionally relevant HRD evident at genomic and transcriptomic levels pointing to genomic instability in breast cancer brain metastasis which is of potential translational significance.
| S-EPMC8791982 | biostudies-literature
Project description:BackgroundSynthetic lethality describes a genetic interaction between two perturbations, leading to cell death, whereas neither event alone has a significant effect on cell viability. This concept can be exploited to specifically target tumor cells. CRISPR viability screens have been widely employed to identify cancer vulnerabilities. However, an approach to systematically infer genetic interactions from viability screens is missing.MethodsHere we describe PAn-canceR Inferred Synthetic lethalities (PARIS), a machine learning approach to identify cancer vulnerabilities. PARIS predicts synthetic lethal (SL) interactions by combining CRISPR viability screens with genomics and transcriptomics data across hundreds of cancer cell lines profiled within the Cancer Dependency Map.ResultsUsing PARIS, we predicted 15 high confidence SL interactions within 549 DNA damage repair (DDR) genes. We show experimental validation of an SL interaction between the tumor suppressor CDKN2A, thymidine phosphorylase (TYMP) and the thymidylate synthase (TYMS), which may allow stratifying patients for treatment with TYMS inhibitors. Using genome-wide mapping of SL interactions for DDR genes, we unraveled a dependency between the aldehyde dehydrogenase ALDH2 and the BRCA-interacting protein BRIP1. Our results suggest BRIP1 as a potential therapeutic target in ~ 30% of all tumors, which express low levels of ALDH2.ConclusionsPARIS is an unbiased, scalable and easy to adapt platform to identify SL interactions that should aid in improving cancer therapy with increased availability of cancer genomics data.
| S-EPMC8401190 | biostudies-literature
Project description:High-grade serous ovarian carcinoma (HGSOC) is a highly aggressive and intractable neoplasm, mainly because of its rapid dissemination into the abdominal cavity, a process that is favored by tumor-associated peritoneal ascites. The precise molecular alterations involved in HGSOC onset and progression remain largely unknown due to the high biological and genetic heterogeneity of this tumor. We established a set of different tumor samples (termed the As11-set) derived from a single HGSOC patient, consisting of peritoneal ascites, primary tumor cells, ovarian cancer stem cells (OCSC) and serially propagated tumor xenografts. The As11-set was subjected to an integrated RNA-seq and DNA-seq analysis which unveiled molecular alterations that marked the different types of samples. Our profiling strategy yielded a panel of signatures relevant in HGSOC and in OCSC biology. When such signatures were used to interrogate the TCGA dataset from HGSOC patients, they exhibited prognostic and predictive power. The molecular alterations also identified potential vulnerabilities associated with OCSC, which were then tested functionally in stemness-related assays. As a proof of concept, we defined PI3K signaling as a novel druggable target in OCSC.
| S-EPMC9310611 | biostudies-literature
Project description:Complicated structures consisting of multi-layers with a multi-modal array of device components, i.e., so-called patterned multi-layers, and their corresponding circuit designs for signal readout and addressing are used to achieve a macroscale electronic skin (e-skin). In contrast to this common approach, we realized an extremely simple macroscale e-skin only by employing a single-layered piezoresistive MWCNT-PDMS composite film with neither nano-, micro-, nor macro-patterns. It is the deep machine learning that made it possible to let such a simple bulky material play the role of a smart sensory device. A deep neural network (DNN) enabled us to process electrical resistance change induced by applied pressure and thereby to instantaneously evaluate the pressure level and the exact position under pressure. The great potential of this revolutionary concept for the attainment of pressure-distribution sensing on a macroscale area could expand its use to not only e-skin applications but to other high-end applications such as touch panels, portable flexible keyboard, sign language interpreting globes, safety diagnosis of social infrastructures, and the diagnosis of motility and peristalsis disorders in the gastrointestinal tract.
| S-EPMC5593817 | biostudies-literature
Project description:Although deep learning (DL) has demonstrated impressive diagnostic performance for a variety of computational pathology tasks, this performance often markedly deteriorates on whole slide images (WSI) generated at external test sites. This phenomenon is due in part to domain shift, wherein differences in test-site pre-analytical variables (e.g., slide scanner, staining procedure) result in WSI with notably different visual presentations compared to training data. To ameliorate pre-analytic variances, approaches such as CycleGAN can be used to calibrate visual properties of images between sites, with the intent of improving DL classifier generalizability. In this work, we present a new approach termed Multi-Site Cross-Organ Calibration based Deep Learning (MuSClD) that employs WSIs of an off-target organ for calibration created at the same site as the on-target organ, based off the assumption that cross-organ slides are subjected to a common set of pre-analytical sources of variance. We demonstrate that by using an off-target organ from the test site to calibrate training data, the domain shift between training and testing data can be mitigated. Importantly, this strategy uniquely guards against potential data leakage introduced during calibration, wherein information only available in the testing data is imparted on the training data. We evaluate MuSClD in the context of the automated diagnosis of non-melanoma skin cancer (NMSC). Specifically, we evaluated MuSClD for identifying and distinguishing (a) basal cell carcinoma (BCC), (b) in-situ squamous cell carcinomas (SCC-In Situ), and (c) invasive squamous cell carcinomas (SCC-Invasive), using an Australian (training, n = 85) and a Swiss (held-out testing, n = 352) cohort. Our experiments reveal that MuSCID reduces the Wasserstein distances between sites in terms of color, contrast, and brightness metrics, without imparting noticeable artifacts to training data. The NMSC-subtyping performance is statistically improved as a result of MuSCID in terms of one-vs. rest AUC: BCC (0.92 vs 0.87, p = 0.01), SCC-In Situ (0.87 vs 0.73, p = 0.15) and SCC-Invasive (0.92 vs 0.82, p = 1e-5). Compared to baseline NMSC-subtyping with no calibration, the internal validation results of MuSClD (BCC (0.98), SCC-In Situ (0.92), and SCC-Invasive (0.97)) suggest that while domain shift indeed degrades classification performance, our on-target calibration using off-target tissue can safely compensate for pre-analytical variabilities, while improving the robustness of the model.
| S-EPMC9825103 | biostudies-literature
Project description:BackgroundThe conventional procedure of skin-related disease detection is a visual inspection by a dermatologist or a primary care clinician, using a dermatoscope. The suspected patients with early signs of skin cancer are referred for biopsy and histopathological examination to ensure the correct diagnosis and the best treatment. Recent advancements in deep convolutional neural networks (CNNs) have achieved excellent performance in automated skin cancer classification with accuracy similar to that of dermatologists. However, such improvements are yet to bring about a clinically trusted and popular system for skin cancer detection.ObjectiveThis study aimed to propose viable deep learning (DL) based method for the detection of skin cancer in lesion images, to help physicians in diagnosis.Material and methodsIn this analytical study, a novel DL based model was proposed, in which other than the lesion image, the patient's data, including the anatomical site of the lesion, age, and gender were used as the model input to predict the type of the lesion. An Inception-ResNet-v2 CNN pretrained for object recognition was employed in the proposed model.ResultsBased on the results, the proposed method achieved promising performance for various skin conditions, and also using the patient's metadata in addition to the lesion image for classification improved the classification accuracy by at least 5% in all cases investigated. On a dataset of 57536 dermoscopic images, the proposed approach achieved an accuracy of 89.3%±1.1% in the discrimination of 4 major skin conditions and 94.5%±0.9% in the classification of benign vs. malignant lesions.ConclusionThe promising results highlight the efficacy of the proposed approach and indicate that the inclusion of the patient's metadata with the lesion image can enhance the skin cancer detection performance.
| S-EPMC9759648 | biostudies-literature
Project description:Parkinson's disease (PD) is the second most common neurodegenerative disease; gait impairments are typical and are associated with increased fall risk and poor quality of life. Gait is potentially a useful biomarker to help discriminate PD at an early stage, however the optimal characteristics and combination are unclear. In this study, we used machine learning (ML) techniques to determine the optimal combination of gait characteristics to discriminate people with PD and healthy controls (HC). 303 participants (119 PD, 184 HC) walked continuously around a circuit for 2-minutes at a self-paced walk. Gait was quantified using an instrumented mat (GAITRite) from which 16 gait characteristics were derived and assessed. Gait characteristics were selected using different ML approaches to determine the optimal method (random forest with information gain and recursive features elimination (RFE) technique with support vector machine (SVM) and logistic regression). Five clinical gait characteristics were identified with RFE-SVM (mean step velocity, mean step length, step length variability, mean step width, and step width variability) that accurately classified PD. Model accuracy for classification of early PD ranged between 73-97% with 63-100% sensitivity and 79-94% specificity. In conclusion, we identified a subset of gait characteristics for accurate early classification of PD. These findings pave the way for a better understanding of the utility of ML techniques to support informed clinical decision-making.
| S-EPMC6872822 | biostudies-literature