Verbal communication is the most prevalent mode of communication for people to express themselves and emotions play a significant role in such human interactions. It plays a great role in capturing feedback in various situations in societal communications between humans. A deep learning-based solution for speech emotion detection is much required in the Healthcare domain for treatment monitoring and treatment course correction through patient feedbacks. It is important to know a patient's emotion during and after treatment to understand the effect of medicines and also to get feedback on treatment. Speech Emotion Recognition (SER) systems play a vital role in better understanding of patient emotions. Continuous collection of relevant patient’s experiences and feedback as recorded audio files and carry such analysis using deep learning techniques. The paper proposes development and use of SER based on deep learning techniques for patient emotion detection. The Long Short-Term Memory (LSTM) architecture of deep learning is employed for developing the system. Developed solutions on SER will be helpful for hospitals and doctors to serve patients better and to provide personalized healthcare services including in remote mode and virtual clinics.