Experiencing sensory-pleasing environments and recreating perceptions of past lived environments can benefit patients greatly in coping with irreversible conditions (e.g., dementia) or during healing after traumatic experiences. Such environments can evoke emotional responses that are associated with a patient's positive memories of a place, an activity, or people. A design concept of an intimate architectural space informed by knowledge of anticipatory systems was developed for the purpose of generating variable personalized environments. In particular, such environments are designed for patients, according to their individual demands and expectations. The design is guided by aesthetic data acquired through perceptual user studies. Such adaptive spaces can be delivered to a large number of users in healthcare in the form of Architecture-as-Service (AaS). The research presented expands the notion of AaS and is premised upon an IoT application.