Background Bone marrow cytology is required to make a hematological diagnosis, influencing critical clinical decision points in hematology. However, bone marrow cytology is tedious, limited to experienced reference centers and associated with inter-observer variability. This may lead to a delayed or incorrect diagnosis, leaving an unmet need for innovative supporting technologies. Methods We develop an end-to-end deep learning-based system for automated bone marrow cytology. Starting with a bone marrow aspirate digital whole slide image, our system rapidly and automatically detects suitable regions for cytology, and subsequently identifies and classifies all bone marrow cells in each region. This collective cytomorphological information is captured in a representation called Histogram of Cell Types (HCT) quantifying bone marrow cell class probability distribution and acting as a cytological patient fingerprint. Results Our system achieves high accuracy in region detection (0.97 accuracy and 0.99 ROC AUC), and cell detection and cell classification (0.75 mean average precision, 0.78 average F1-score, Log-average miss rate of 0.31). Conclusions HCT has potential to eventually support more efficient and accurate diagnosis in hematology, supporting AI-enabled computational pathology.

The number and size of medical databases are rapidly increasing, and the advanced models of data mining techniques could help physicians to make efficient and applicable decisions. The challenges of heart disease data include the feature selection, the number of the samples; imbalance of the samples, lack of magnitude for some features, etc. This study mainly focuses on the feature selection improvement and decreasing the numbers of the features. In this study, imperialist competitive algorithm with meta-heuristic approach is suggested in order to select prominent features of the heart disease. This algorithm can provide a more optimal response for feature selection toward genetic in compare with other optimization algorithms. Also, the K-nearest neighbor algorithm is used for the classification. Evaluation result shows that by using the proposed algorithm, the accuracy of feature selection technique has been improved.

Recognizing potential sites for landfill has increasingly become an important waste management strategy around the world. This study aims to determine municipal solid waste (MSW) landfills in SaharKhiz Region located in Gilan by comparing Fuzzy logic and Boolean logic. Fuzzy logic, which has been used in this research, is based on weighted linear combination (WLC); however, the utilized Boolean logic is considered only to determine the accuracy and validity of every method. At first, the Boolean logic using a geographical information system (GIS) is used to recognize potential and excluded zones, based on zero and one value system. In the next phase, Fuzzy logic is used, between zero and one, to standardize information layers, based on their type (increasing or decreasing). The final weight of every layer was determined using the analytical hierarchy process. Finally, the WLC method was used to integrate layers in the GIS environment to provide the final site suitability map in five classes of Fuzzy membership degree. The results show that Fuzzy logic, based on WLC, has more flexibility to resolve conflicts of human judgment, and it also has higher accuracy than Boolean logic in the selection of optimal landfill sites for MSW in SaharKhiz Region, Gilan Province, based on ecological and socioeconomic parameters.

Background Pathology synopses consist of semi-structured or unstructured text summarizing visual information by observing human tissue. Experts write and interpret these synopses with high domain-specific knowledge to extract tissue semantics and formulate a diagnosis in the context of ancillary testing and clinical information. The limited number of specialists available to interpret pathology synopses restricts the utility of the inherent information. Deep learning offers a tool for information extraction and automatic feature generation from complex datasets. Methods Using an active learning approach, we developed a set of semantic labels for bone marrow aspirate pathology synopses. We then trained a transformer-based deep-learning model to map these synopses to one or more semantic labels, and extracted learned embeddings (i.e., meaningful attributes) from the model’s hidden layer. Results Here we demonstrate that with a small amount of training data, a transformer-based natural language model can extract embeddings from pathology synopses that capture diagnostically relevant information. On average, these embeddings can be used to generate semantic labels mapping patients to probable diagnostic groups with a micro-average F1 score of 0.779 Â ± 0.025. Conclusions We provide a generalizable deep learning model and approach to unlock the semantic information inherent in pathology synopses toward improved diagnostics, biodiscovery and AI-assisted computational pathology.