Computer-aided diagnosis (CAD) systems can help radiologists in numerous medical tasks including classification and staging of the various diseases. The 3D tomosynthesis imaging technique adds value ...to the CAD systems in diagnosis and classification of the breast lesions. Several convolutional neural network (CNN) architectures have been proposed to classify the lesion shapes to the respective classes using a similar imaging method. However, not only is the black box nature of these CNN models questionable in the healthcare domain, but so is the morphological-based cancer classification, concerning the clinicians. As a result, this study proposes both a mathematically and visually explainable deep-learning-driven multiclass shape-based classification framework for the tomosynthesis breast lesion images. In this study, authors exploit eight pretrained CNN architectures for the classification task on the previously extracted regions of interests images containing the lesions. Additionally, the study also unleashes the black box nature of the deep learning models using two well-known perceptive explainable artificial intelligence (XAI) algorithms including Grad-CAM and LIME. Moreover, two mathematical-structure-based interpretability techniques, i.e., t-SNE and UMAP, are employed to investigate the pretrained models’ behavior towards multiclass feature clustering. The experimental results of the classification task validate the applicability of the proposed framework by yielding the mean area under the curve of 98.2%. The explanability study validates the applicability of all employed methods, mainly emphasizing the pros and cons of both Grad-CAM and LIME methods that can provide useful insights towards explainable CAD systems.
Artificial intelligence algorithms have become extensively utilized in survival analysis for high-dimensional, multi-source data. However, due to their complexity, these methods often yield poorly ...interpretable outcomes, posing challenges in the analysis of several conditions. One of these conditions is obstructive sleep apnea, a sleep disorder characterized by the simultaneous occurrence of comorbidities. Survival analysis provides a potential solution for assessing and categorizing the severity of obstructive sleep apnea, aiding personalized treatment strategies. Given the critical role of time in such scenarios and considering limitations in model interpretability, time-dependent explainable artificial intelligence algorithms have been developed in recent years for direct application to basic Machine Learning models, such as Cox regression and survival random forest. Our work aims to enhance model selection in OSA survival analysis using time-dependent XAI for Machine Learning and Deep Learning models. We developed an end-to-end pipeline, training several survival models and selecting the best performers. Our top models—Cox regression, Cox time, and logistic hazard—achieved good performance, with C-index scores of 0.81, 0.78, and 0.77, and Brier scores of 0.10, 0.12, and 0.11 on the test set. We applied SurvSHAP methods to Cox regression and logistic hazard to investigate their behavior. Although the models showed similar performance, our analysis established that the results of the log hazard model were more reliable and useful in clinical practice compared to those of Cox regression in OSA scenarios.
In recent years, a constant and fast information growing has characterized digital applications in the majority of real-life scenarios. Thus, a new information asset, namely Big Data, has been ...defined and lead to different challenges, mainly related to data storage, management and analysis. Focusing on the last challenge, several Big Data analytics techniques have been developed, based on Machine Learning and Deep Learning paradigms. When dealing with Big Data, traditional approaches often take a lot of time to produce even a single predictive model, due to the extremely high demand of computational resources.
The design of approaches specifically oriented to Big Data is required to overcome these computational issues. Most solutions rely on the deployment of Big Data analytics infrastructures on a cluster of machines and/or on parallelization techniques. When deployment and parallelization apply to Machine Learning and Deep Learning, we can refer to the terms Distributed Machine Learning and Distributed Deep Learning, respectively.
We here discuss the main principles and features of Distributed Machine Learning and Distributed Deep Learning frameworks. The main contribution of this work is a survey of solutions proposed in the literature, through the investigation of selected features and capabilities. In particular, the survey provides a comparative analysis according to the following classification criteria: implemented parallelization technique, supporting device, supported architecture, implemented communication mode, working mode, and class of algorithms.
The paper also gives an overview of the most commonly used criteria and metrics for the performance evaluation of analyzed frameworks; finally, some emerging but promising optimization techniques are reviewed apart from our classification.
•Big Data analytics leads to challenges related to data management and analysis.•An analysis of the related solutions proposed in the literature is required.•The analysis is focused on Distributed Machine Learning and Deep Learning.•Overview of framework performance evaluation criteria with optimization techniques.
To test inter-fraction reproducibility, intrafraction stability, technician aspects, and patient/physician's comfort of a dedicated immobilization solution for Brain Linac-based radiation therapy ...(RT).
A pitch-enabled head positioner with an open-face mask were used and, to evaluate inter- and intrafraction variations, 1-3 Cone-Beam Computed Tomography (CBCT) were performed. Surface Guided Radiation Therapy (SGRT) was used to evaluate intrafraction variations at 3 time points: initial (i), final (f), and monitoring (m) (before, end, and during RT). Data regarding technician mask aspect were collected.
Between October 2019 and April 2020, 69 patients with brain disease were treated: 45 received stereotactic RT and 24 conventional RT; 556 treatment sessions and 863 CBCT's were performed. Inter-fraction CBCT mean values were longitudinally 0.9 mm, laterally 0.8 mm, vertically 1.1 mm, roll 0.58°, pitch 0.59°, yaw 0.67°. Intrafraction CBCT mean values were longitudinally 0.3 mm, laterally 0.3 mm, vertically 0.4 mm, roll 0.22°, pitch 0.33°, yaw 0.24°. SGRT intrafraction mean values were: i_, m_, f_ longitudinally 0.09 mm, 0.45 mm, 0.31 mm; i_, m_, f_ laterally 0.07 mm, 0.36 mm, 0.20 mm; i_, m_, f_ vertically 0.06 mm, 0.31 mm, 0.22 mm; i_, m_, f_ roll 0.025°, 0.208°, 0.118°; i_, m_, f_ pitch 0.036°, 0.307°, 0.194°; i_, m_, f_ yaw 0.039°, 0.274°, 0.189°.
This immobilization solution is reproducible and stable. Combining CBCT and SGRT data confirm that 1 mm CTV-PTV margin for Linac-based SRT was adequate. Using open-face mask and SGRT, for conventional RT, radiological imaging could be omitted.
•The Oxford Classification is the most successful nephropathology scoring system.•An end-to-end deep learning pipeline was developed to replicate expert scoring.•Segmentation and classification ...models are freely available to the community.
The Oxford Classification for IgA nephropathy is the most successful example of an evidence-based nephropathology classification system. The aim of our study was to replicate the glomerular components of Oxford scoring with an end-to-end deep learning pipeline that involves automatic glomerular segmentation followed by classification for mesangial hypercellularity (M), endocapillary hypercellularity (E), segmental sclerosis (S) and active crescents (C).
A total number of 1056 periodic acid–Schiff (PAS) whole slide images (WSIs), coming from 386 kidney biopsies, were annotated. Several detection models for glomeruli, based on the Mask R-CNN architecture, were trained on 587 WSIs, validated on 161 WSIs, and tested on 127 WSIs. For the development of segmentation models, 20,529 glomeruli were annotated, of which 16,571 as training and 3958 as validation set. The test set of the segmentation module comprised of 2948 glomeruli. For the Oxford classification, 6206 expert-annotated glomeruli from 308 PAS WSIs were labelled for M, E, S, C and split into a training set of 4298 glomeruli from 207 WSIs, and a test set of 1908 glomeruli. We chose the best-performing models to construct an end-to-end pipeline, which we named MESCnn (MESC classification by neural network), for the glomerular Oxford classification of WSIs.
Instance segmentation yielded excellent results with an AP50 ranging between 78.2–80.1 % (79.4 ± 0.7 %) on the validation and 75.1–77.7 % (76.5 ± 0.9 %) on the test set. The aggregated Jaccard Index was between 73.4–75.9 % (75.0 ± 0.8 %) on the validation and 69.1–73.4 % (72.2 ± 1.4 %) on the test set. At granular glomerular level, Oxford Classification was best replicated for M with EfficientNetV2-L with a mean ROC-AUC of 90.2 % and a mean precision/recall area under the curve (PR-AUC) of 81.8 %, best for E with MobileNetV2 (ROC-AUC 94.7 %) and ResNet50 (PR-AUC 75.8 %), best for S with EfficientNetV2-M (mean ROC-AUC 92.7 %, mean PR-AUC 87.7 %), best for C with EfficientNetV2-L (ROC-AUC 92.3 %) and EfficientNetV2-S (PR-AUC 54.7 %). At biopsy-level, correlation between expert and deep learning labels fulfilled the demands of the Oxford Classification.
We designed an end-to-end pipeline for glomerular Oxford Classification on both a granular glomerular and an entire biopsy level. Both the glomerular segmentation and the classification modules are freely available for further development to the renal medicine community.
ARGO 2.0: a Hybrid NLP/ML Framework for Diagnosis Standardization Berloco, Francesco; Ciavarella, Sabino; Colucci, Simona ...
2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC),
01/2023, Letnik:
2023
Conference Proceeding, Journal Article
A relevant problem in medicine is the standardization of the diagnosis associated with a clinical case. Although diagnosis formulation is an intrinsically subjective and uncertain process, its ...standardization may take benefit from digital solutions automating the routines at the basis of such a decision. In this work, we propose ARGO 2.0: a framework for the development of decision support systems for diagnosis formulation. The framework can read free-text reports and store their clinically relevant information as personalized electronic Case Report Forms. A hybrid strategy, exploiting the synergy of Natural Language Processing and Machine Learning techniques, is used to automatically suggest a diagnosis in a standardized fashion. ARGO 2.0 has been designed to be template-independent and easily tailored to specific medical fields. We here demonstrate its feasibility in hemo lympho-pathology, by detailing its implementation, object of an ongoing validation campaign in a standing medical institute. ARGO 2.0 achieved an average Accuracy of 95.07%, an average precision of 94.85%, an average Recall of 96.31% and a F-Score of 95.32% onto the test set, outperforming both its embedded components, based on Natural Language Processing and Machine Learning.
AIM: To investigate H pylori expression in gastric cancer patients in relation to primary tumor angiogenic markers, such as microvessel density (MVD), thymidine phosphorylase (TP), vascular ...endothelial growth factor receptor-1 (VEGF-R1), p53 and circulating VEGF levels.
METHODS: Angiogenic markers were analyzed immunohistochemically in 56 primary gastric cancers. H pylori cytotoxin (vacA) and the cytotoxin-associated gene (cagA) amplification were evaluated using PCR assay. Serum H pylori IgG antibodies and serum/plasma circulating VEGF levels were detected in 39 and 38 patients by ELISA, respectively.
RESULTS: A total of 69% of patients were positive for circulating IgG antibodies against Hpylori. cagA-positive H pylor/strains were found in 41% of gastric patients. vacA was found in 50% of patients; s1 strains were more highly expressed among vacA-positive patients. The presence of the s1 strain was significantly associated with cagA (P = 0.0001). MVD was significantly correlated with both tumor VEGF expression (r = 0.361, P = 0.009) and serum VEGF levels (r = -0.347, P = 0.041).Conversely, neither VEGF-R1 expression nor MVD was related to p53 expression. However, H pylori was not related to any angiogenic markers except for the plasma VEGF level (P = 0.026).
CONCLUSION: H pylori antigen is related to higher plasma VEGF levels, but not to angiogenic characteristics. It can be hypothesized that the toxic effects of H pylori on angiogenesis occurs in early preclinical disease phase or in long-lasting aggressive infections, but only when high H pylori IgG levels are persistent.
Spermatogonial kinetics in humans Di Persio, Sara; Saracino, Rossana; Fera, Stefania ...
Development,
10/2017, Letnik:
144, Številka:
19
Journal Article
Recenzirano
Odprti dostop
The human spermatogonial compartment is essential for daily production of millions of sperm. Despite this crucial role, the molecular signature, kinetic behavior and regulation of human spermatogonia ...are poorly understood. Using human testis biopsies with normal spermatogenesis and by studying marker protein expression, we have identified for the first time different subpopulations of spermatogonia. MAGE-A4 marks all spermatogonia, KIT marks all B spermatogonia and UCLH1 all Apale-dark (Ap-d) spermatogonia. We suggest that at the start of the spermatogenic lineage there are Ap-d spermatogonia that are GFRA1
, likely including the spermatogonial stem cells. Next, UTF1 becomes expressed, cells become quiescent and GFRA1 expression decreases. Finally, GFRA1 expression is lost and subsequently cells differentiate into B spermatogonia, losing UTF1 and acquiring KIT expression. Strikingly, most human Ap-d spermatogonia are out of the cell cycle and even differentiating type B spermatogonial proliferation is restricted. A novel scheme for human spermatogonial development is proposed that will facilitate further research in this field, the understanding of cases of infertility and the development of methods to increase sperm output.
To date, in the human seminiferous epithelium, only six associations of cell types have been distinguished, subdividing the epithelial cycle into six stages of very different duration. This hampers ...comparisons between studies on human and laboratory animals in which the cycle is usually subdivided into 12 stages. We now propose a new stage classification on basis of acrosomal development made visible by immunohistochemistry (IHC) for (pro)acrosin. IHC for acrosin gives results that are comparable to periodic acid Schiff staining. In the human too, we now distinguish 12 stages that differ from each other in duration by a factor of two at most. B spermatogonia are first apparent in stage I, preleptotene spermatocytes are formed in stage V, leptonema starts in stage VII, and spermiation takes place at the end of stage VI. A similar timing was previously observed in several monkeys. Stage identification by way of IHC for acrosin appeared possible for tissue fixed in formalin, Bouin fixative, diluted Bouin fixative, Cleland fluid, and modified Davidson fixative, indicating a wide applicability. In addition, it is also possible to distinguish the 12 stages in glutaraldehyde/osmium-tetroxide fixed/plastic embedded testis material without IHC for acrosin. The new stage classification will greatly facilitate research on human spermatogenesis and enable a much better comparison with results from work on experimental animals than hitherto possible. In addition, it will enable a highly focused approach to evaluate spermatogenic impairments, such as germ cell maturation arrests or defects, and to study details of germ cell differentiation.