OUR CASE STUDIES
Artificial intelligence and perimeter security
Recent advances in Computer Vision allow real-time recognition of moving objects within complex scenes. These innovations now find applications in many areas including autonomous driving, vehicle traffic management and various industrial and security applications. Sacbo, in partnership with SORINT.tek, has equipped itself with a system that identifies entry into user-definable areas specific to each person
Prognostics and Health Management
An overview about Prognostics and Health Management, an engineering discipline that aims to the real-time health assessment of a system and the prediction of its future state. It is tied to the concept of Condition-Based Maintenance (CBM), a management philosophy whose purpose is the optimization of the maintenance activity in order to reduce downtimes and
Jupyter (R)Evolution for Data Engineering
The Jupyter Project revolutionised the Data Science field, speeding up the exploration of data. After a progressive approach, another curious phenomenon occurred: even the area of Data Engineering improved with the introduction of Jupyter. The deployment of Jupyter in the Infocert’s Data Engineering team and the benefits gained as soon as it became the ETL
Deep Learning applied at medical domain
TAC scans are an important diagnostic tool for the screening, diagnosis and evaluation of Covid-19 induced pneumonia cases. Poliambulanza Brescia and SORINT.tek are developing a system for the collection of CT scans and their automatic annotation to identify regions with lesions associated with the presence of pneumonia. The system can be used for case screening
NLP & ChatBot with Rasa
A small introduction to the world of chatbots and NLP. An excellent opportunity for anyone who is interested in the subject, but does not know where to start. After a brief introduction to the world of chatbots and NLP, the basic concepts will be presented, and finally the use of Rasa to create a simple
Machine Learning Pipelines with Sklearn
The Sklearn module Pipeline is a handy tool to chain processing and prediction step together in a whole block. It makes Data Scientist life easier. A brief introduction of the module and its methods together with few practical examples.
Energy management & Smart metering: from infrastructure to data management
An asset management and operation intelligence platform is one of the key components of the path to digitization, to speed up data acquisition, processing and analysis processes and to translate the Internet of Things into business value. A concrete case of the use of the MangroviaIot solution in partnership with Valtellina for the Iren group
MangroviaIoT – asset management & operation intelligence
From many years of experience in software development and advanced analytics in the Utility & Energy sector was born MangroviaIoT, a modular and multi-level infrastructure that allows to manage and monitor the status of a device infrastructure, acquire, collect, monitor and display sensor data and create a database useful for Advanced Analytics.
Artificial Neural Network Based Predictive Control applied to Residential Buildings
We propose a supervisory ANN-MPC controller for residential buildings, that is able to simultaneously minimize the energy consumption and cost and to maintain the desired thermal comfort, while taking into account occupant expectations and physical constraints like the energy price and the power limitations. The model, applied to a building with an automatic HVAC system
Conditional networks for screening of breast cancer metastases in lymph nodes
Medical Imaging with Deep Learning (MIDL) – London, 8th-10th July 2019 We researched conditional networks to screen Whole Slide Images (WSI) of sentinel lymph-nodes and to segment areas which might contain metastatic lesions. The network output is intended for a subsequent inspection by a human pathologist. The method is the basis for a more interactive
Analytics and gas distribution
A Smart Metering System is an intelligent network through which an advanced and efficient measurement system is able to remotely control and manage end-user counters. It consists of three basic elements – the meter, a communication network and a control center – for collecting information from meters, validate such information and activate the computerized processes
Industry 4.0: Prognostic and Health Management
Tra i molti benefici che la digitalizzazione dell’industria può portare vi è la possibilità di ridurre i costi di manutenzione e incrementare l’affidabilità dei macchinari attraverso la Smart Maintenance. Questo termine spesso usato come sinonimo di Predictive Maintenance indica la capacità di effettuare la manutenzione dei macchinari solo quando effettivamente necessario e quindi minimizzando sia
Slide screening of metastases in lymph nodes with deep learning algorithms
Breast cancer is the most common form of cancer among women in Italy: approximately one woman every eight is diagnosed with breast cancer during her lifetime.
Automated slide screening for the diagnosis of breast cancer metastases in lymph nodes
European Conference on Digital Pathology (ECDP) – Helsinki, 29th May–1st June 2018Development and validation in a clinical environment of a method that can adapt its behavior to the work practices, digital instrumentation, and patient population of the clinic