Nowadays, one of the important aspects of research on call/contact centre (CC) systems is how to automate their operations. Process automation is influenced by the continuous development in the implementation of virtual assistants. The effectiveness of virtual assistants depends on numerous factors. One of the most important is correctly recognizing the intent of clients conversing with the machine. Recognizing intentions is not an easy process, as often the client’s actual intentions can only be correctly identified after considering the client’s emotional state. When it comes to human–machine communication, the ability of a virtual assistant to recognize the client’s emotional state would greatly improve its effectiveness. This paper proposes a new method for recognizing interlocutors’ emotions dedicated directly to contact centre systems. The developed method provides opportunities to determine emotional states in text and voice channels. It provides opportunities to explore both the client’s and the agent’s emotional states. Information about agents’ emotions can be used to build their behavioural profiles, which is also applicable in contact centres. In addition, the paper explored the possibility of emotion assessment based on automatic transcriptions of recordings, which also positively affected emotion recognition performance in the voice channel. The research used actual conversations that took place during the operation of a large, commercial contact centre. The proposed solution makes it possible to recognize the emotions of customers contacting the hotline and agents handling these calls. Using this information in practical applications can increase the efficiency of agents’ work, efficiency of bots used in CC and increase customer satisfaction.

Interactions between extracellular matrix (ECM) and epithelial cells are necessary for proper organisation and function of the epithelium. In the present study we show that bovine mammary epithelial cell line BME-UV1 cultured on ECM components, commercially available as Matrigel™, constitutes a good model for studying mechanisms controlling functional differentiation of the bovine mammary gland. In contact with Matrigel BME-UV1 cells induce apicobasal polarity, and within 16 days form three dimensional (3D) acinar structures with a centrally localized hollow lumen, which structurally resemble mammary alveoli present in the functionally active mammary gland. We have shown that the 3D culture system enables a high expression and proper localisation of integrin receptors and tight junction proteins in BME-UV1 cells to be induced. This effect was not obtained in cells grown in the classical 2D culture system on plastic. Moreover, ECM highly stimulated the synthesis of one of the major milk proteins, β-casein, even in the absence of prolactin. Our results show that contact with ECM plays an important role in the lactogenic activity of bovine MECs, however, prolactin is necessary for the efficient secretion of milk proteins.

The article presents the Parrot AI smart system, which supports physicians specialising in internal medicine and pediatrics. The solution is designed to automate time-consuming administrative processes and assist physicians in clinical decision-making. The main purpose of the system is to relieve medical staff of administrative tasks, such as manually filling out medical records, which translates into increased work efficiency and more time to focus on direct contact with patients. Parrot AI uses advanced natural language processing methods and generative language models to enable automated semantic analysis of medical interviews. Based on this, the system generates the content necessary to complete the electronic medical records and assists the physician by suggesting preliminary diagnoses and treatment options. Interaction with the patient can take place both at the initial medical interview stage (via chatbot or voicebot) and during a personal appointment at the clinic. Functional tests have demonstrated that the solution is highly effective. All tested components of the system achieved results well above the 80% effectiveness threshold, confirming the validity of further implementing the system into everyday clinical practice. Parrot AI has the potential not only to streamline physicians’ work, but also to improve the quality and consistency of medical records.

This article presents extensive research and analysis of the location capability of Bluetooth low energy (BLE) in designing real-time location systems (RTLSs). The main assumptions for the proposed approach are to optimize precision of location to less than 2 m, its reach up to 20 m, and system response time not exceeding 5 s. In addition, the system should be characterized by low energy consumption, and low construction and operating costs. In order to meet the above assumptions, extensive studies of received signal strength indicator (RSSI) radio signal parameters and time interval (TI) parameters were carried out. The tests were performed at a distance of 1 to 20 m for different types of emitter (beacon)–scanner pairs (microprogrammable platform). Detailed studies of location accuracy were performed using several nonlinear regression methods. The selection and assessment of the reliability of the approximating models was based on the information criterion measures and cross-validation. The influence of the training set size and the test set size on the adopted model’s reliability was examined. Based on TI values, predicted response time was estimated for real-time localization, and the relationship between system dynamics and localization accuracy was investigated.