Accurate volumetric assessment of soft-tissue tumors is essential for surgical planning and outcome monitoring. Although computed tomography (CT) is the standard, it involves radiation exposure and infrastructural dependency. Smartphone-based light detection and ranging (LiDAR) sensors with neural radiance fields (NeRFs) offer a portable, noninvasive alternative, yet clinical applications remain underexplored. We report the case of a 75-year-old man with a long-standing posterior cervical lipoma. Pre- and postoperative 3-dimensional data were acquired using conventional CT and a smartphone-based LiDAR scanning application with NeRF rendering. Tumor volume was estimated by calculating the difference in the surface volume of the head and neck between the 2 time points. Accuracy, feasibility, and potential sources of error were evaluated. The tumor volume was 156 mL via CT and 145 mL via smartphone 3-dimensional scan, with a minimal error margin of 7.0%. The actual specimen weighed 105 g, showing strong concordance. A slight overestimation occurred due to the inclusion of hair-bearing regions; however, differential volume calculation remained robust. The scan required less than 1 minute and was performed under standard clinical conditions without advanced equipment. This preliminary report demonstrates that smartphone-based LiDAR and NeRF technology enables simple, fast, and noninvasive volumetric assessment of soft-tissue masses with clinically promising accuracy. As these technologies evolve, they hold strong potential for broader application in surgical planning, outpatient follow-up, and remote care, particularly in resource-limited settings. This study provides early evidence for integration into future clinical workflows.
Nakaso et al. (Fri,) studied this question.