Gas-filled microbubbles (MBs) can be locally cavitated by focused ultrasound (FUS) to trigger stable cavitation (SC) and inertial cavitation (IC) to induce blood–brain barrier (BBB) opening [1,2]. Because of the hemolysis caused by IC [3], SC is a more preferable effect to be conducted to BBB opening. However, to real-time observe the SC effect by MRI is not discussed thoroughly. In this study, we aim to real-time monitor the signal changes of MBs at the present of SC effect in an in vitro agarose phantom by half-Fourier acquisition single-shot turbo spin-echo (HASTE) sequence.While SC effect occurred, the vibration of MBs may lead to flowing effect and MR signals are unable to be refocused in HASTE images. We assumed reduced MR signals may reflect the degree of induced flowing effect. The diluted MBs solutions (lipid shell with C3F8, mean diameters=1.25 µm, concentration = (4.36±0.32)×1010 droplets/mL) [4] were injected into an agarose phantom with two hollow chambers (diameter = 6 mm).The MBs were diluted to 0.001X. A homemade single-element focused piezoelectric transducer was used as the source of FUS sonication. FUS pulses with different acoustic pressure (0.1, 0.2, 0.3 MPa), pulse repetitive frequency (PRF = 1, 10 Hz), and duty cycle (1%, 10%) were applied to observe the SC effect on MR signal changes during FUS transmission. All experiments were conducted in a 3 Tesla MR scanner (Tim Trio, Siemens, Erlangen, Germany). The HASTE sequence (TR/TE= 1270/49 ms, pixel size= 0.78x0.78x3 mm3, pixel bandwidth= 454 Hz/pixel, temporal resolution=1.27 s, and 126 measurements (160 s)) was performed. All images were acquired at the focal plane and were perpendicular to the direction of ultrasound beams. To evaluate changes of signal intensity (SI), a region-of-interesting (ROI) was selected manually with a self-developed program at chamber of MBs (Fig. 1). The SI within ROI was normalized to mean SI before turning-on of FUS pulses (pre-FUS): normalized SI = (SI/SI_(pre-FUS))x100%.Figure 2a displayed typical time courses of normalized SI with PRF=1Hz and duty cycle=1% for different acoustic pressures. Specific statuses during the time courses were marked as: (I) Pre-FUS, (II) signal changes with fluctuation, and (III) Post-FUS. At Pre- and Post-FUS, normalized SI is approximate 100%. During FUS transmitting, the normalized SI changed from 100% to 99.3%, 94.4%, and 88.6% for 0.1, 0.2, and 0.3 MPa FUS pulses, respectively, indicating SI changes increased with increase of acoustic pressure. Figures 2b-2d compared SI changes with various PRF and duty cycle under identical acoustic pressure. With acoustic pressure of 0.1 or 0.2 MPa, PRF= 1 or 10 Hz showed similar time courses whereas SI changes was higher n 1 Hz than 10 Hz if pressure increased to 0.3 MPa. If the PRF fixed as 1 Hz, duty cycle=10% demonstrated larger SI changes than duty cycle=1%. Table 1 summarizes SI changes of all experiments. Figure 3 showed SI changes of experiments with PRF=1Hz and duty cycle=1% under acoustic pressure of 0.2, 0.3, 0.4, 0.8 MPa. The slope of SI increased with acoustic pressure during FUS transmission, exploring that higher acoustic pressure led to faster disruption of MBs 0.8MPa FUS pulses, the IC effect could occur, leading to MBs diminished completely and thus the normalized SI recovered to 100% before the end of FUS transmission.In this study, HASTE sequence was performed to monitor the SI changes at the present of MBs SC effect. We investigated the SI changes of experiments with various conditions of acoustic pressure, PRF, and duty cycle. Upon the transmission of FUS pulses, the normalized SI dropped suddenly and fluctuated till the end of transmission (Fig. 2). Under fixed acoustic pressure and PRF, the longer duty cycle allowed more time to accumulate the energy from FUS pulse and therefore resulted in more substantial flowing effect, accompanying with more SI changes. Regarding to higher PRF, it could be seen as separate the transmission into several sections, leading to less energy accumulation and thus less SI changes, particular in higher acoustic pressure of 0.3 MPa in Fig. 2d. In conclusion, we used HASTE sequence to real-time monitor SI changes at the present of SC effect of MBs. Further studies shall be conducted to clarify the SC effect on SI changes to optimize MR scanning parameters and to comprehend the influence of various MB concentrations and FUS conditions in the future.