Low-Intensity Pulsed Ultrasound Stimulation (LIPUS) is a therapeutic modality used for bone tissue regeneration and healing. Its clinical efficacy is still debated, as the underlying physical phenomena remain poorly understood. The interaction between ultrasonic waves and cells, likely to trigger mechanotransduction inducing bone regeneration, is at the center of scientific concerns on the subject. In order to get new insights into these phenomena, the development of in vitro experiments is a key step but special attentions should be paid concerning to the actual acoustic area covered that has to be sufficiently large and homogeneous. To address this issue, an acoustic lens can be placed on the transducer to improve the homogeneity of the acoustic field over the entire cell culture area. A computational model is developed to test several shapes and heights of acoustic lenses and compare their effectiveness in order to find a compromise between the surface covered, the homogeneity of the intensity distribution and the acoustic pressure loss. All the lenses studied improve the enlargement of the field and its homogeneity but they all generate pressure acoustic loss. The best performing lens in terms of field homogeneity is the one that minimizes pressure acoustic loss but covers only 22% of the target surface. The best enlargement (68% of the surface covered) is obtained for a lens that produces a field that is 4 times less homogeneous and 3 times less efficient in terms of pressure acoustic loss. As no one lens is ideal, the choice of the lens should be the result of a compromise taking into account the prioritization of criteria.