In order to compete for a prominent market share, network operators and service providers should retain and increase the verticals' subscription, catering to their needs in order to differentiate themselves from competitors. In this scenario, verticals' satisfaction arises of paramount importance. As such, user experience is becoming a reliable indicator for service providers and telecommunication operators to convey overall end-to-end system functioning. To properly estimate end user satisfaction, operators and service providers require efficient means for quality monitoring and estimation at all layers, in conjunction with mechanisms able to maintain said quality at optimum levels. Given these factors, this paper proposes a mechanism for Quality of Perception (QoP) estimation in e-Health services, enabling the QoP-aware management of network slices fulfilling the requirements of supported services. To this end, the paper proposes a cognitive-based architecture which allows for the collection and monitoring of verticals' data to estimate QoP and provides mechanisms to re-configure the underlying network slices according to the monitored quality levels. A machine learning (ML) model is introduced that aims to forecast any future degradation in the quality perceived by vertical users. In case of a predicted degradation, the proposed architecture reacts and triggers the necessary remedial actions, referred as actuations. In order to evaluate the developed ML model and to showcase the