Focused Ultrasound (HIFU) and Radiofrequency (RF) — They Are Not Competing Technologies

The point of this table is that “what sets the depth” differs between them. In radiofrequency, depth is set by electrode geometry rather than polarity — the presence of a return pad determines only whether current passes through the body, not the current distribution a few millimetres beneath the surface. In actual data the figures were: monopolar spherical electrode 4.37 mm (measured 4.69 ± 0.96), monopolar flat square 10.58 mm (measured 6.80 ± 1.63), bipolar at 1 MHz 20.15 mm and at 0.5 MHz 25.00 mm (J Biomed Sci Eng 2024 — all authors were employees of the device manufacturer). Conversely, even monopolar penetrates shallowly with a small electrode — a 0.5 mm radius electrode reaches only about 140 µm.

In focused ultrasound, the depth is fixed per transducer. Rather than changing depth freely during treatment, the method is to select a transducer of the intended depth. We do not assert focal depth figures for any specific device in this article — they differ by device, and in some cases we could not verify the official specification for the product.

The misconception we hear most often in consultation is “HIFU goes deeper, so it is stronger.” There are two problems with that sentence.

First, depth is set by the settings, not by the technology. As the figures above show, radiofrequency has configurations that calculate to over 20 mm depending on electrode arrangement, and conversely reaches only 140 µm with a small electrode. The generalisation “RF is shallow and HIFU is deep” does not hold.

Second, deeper does not mean better. Where the structure you want to address sits comes first. Raising the density of the dermis as a whole and creating contraction points at a particular layer are different aims.

For the same reason, the explanation that frequency sets the depth is only half right. In 2025 data combining computer simulation with porcine histology, 2 MHz heated the fat layer broadly and deeply while 6.78 MHz heated focally along the fibrous septa — frequency changes where the heat concentrates rather than the depth.

“This principle is established” and “this device produces that result” are entirely different sentences. The principle behind both technologies is old, but the human data for individual devices is mostly a single arm of about 20 people.

With no direct comparison data, we divide by the layer of the problem.

We will note once more that injectables and devices treat different layers. “My skin is thin and dull” and “my facial contour has collapsed” are different problems, and the two often occur together. When they do, we plan the order separately — set out in booster first or lifting first.

Treatments that address collagen with heat show up in two strands — immediate contraction right after delivery and remodelling over weeks to months. The first subsides quickly and what lasts is the second, so it is difficult to judge the result from the face the day after treatment.

In an assessor-blinded study of 20 Korean women, radiofrequency was not significant on clinician assessment at 4 weeks but was at 12 weeks (instrument-measured elasticity was significant from 4 weeks, p < 0.01), with a peak at 4–6 months. Judgement timing in general is set out in when improvement shows, and when to judge it.

Written and reviewed by Lee Chi-Hak, MD, medical director of Miso Clinic in Daegu, South Korea. Every study cited above is given together with its design, its size and the limitations the authors themselves recorded, and where we could not find data, we have said that we could not find any.