Three-dimensional ultrasound fusion imaging in precise needle placement for thermal ablation of hepatocellular carcinoma

Abstract

Purpose

To investigate the value of three-dimensional ultrasound fusion imaging (3DUS FI) technique for guiding needle placement in hepatocellular carcinoma (HCC) thermal ablation.

Methods

A total of 57 patients with 60 HCCs with 3DUS FI-guided thermal ablation were retrospectively included in the study. 3DUS volume data of liver were acquired preoperatively by freehand scanning with the tumor and predetermined 5 mm ablative margin automatically segmented. Plan of needle placement was made through a predetermined simulated ablation zone to ensure a 5 mm ablative margin with the coverage rate toward tumor and ablative margin. With real-time ultrasound and 3DUS fusion imaging, ablation needles were placed according to the plan. After ablation, the ablative margin was immediately evaluated by contrast-enhanced ultrasound and 3DUS fusion imaging. The rate of adequate ablative margin, complete response (CR), local tumor progression (LTP), disease-free survival (DFS), and overall survival (OS) was evaluated.

Results

According to postoperative contrast-enhanced CT or MR imaging, the complete response rate was 100% (60/60), and 83% of tumors (30/36) achieved adequate ablative margin (>5 mm) three-dimensionally. During the follow-up period of 6.0-42.6 months, LTP occurred in 5 lesions, with 1- and 2-year LTP rates being 7.0% and 9.4%. The 1- and 2-year DFS rates were 76.1% and 65.6%, and 1- and 2-year OS rates were 98.1% and 94.0%. No major complications or ablation-related deaths were observed in any patients.

Conclusions

Three-dimensional ultrasound fusion imaging technique may improve the needle placement of thermal ablation for HCC and reduce the rate of LTP.

Keywords:

• Hepatocellular carcinoma
• ablation techniques
• ultrasonography
• fusion image

Introduction

Thermal ablation has been an important option for hepatocellular carcinoma (HCC) treatment, especially for those in early or very early stage (BCLC 0-A stage) but unsuitable for surgical resection, which provides similar survival outcomes and lower rate of adverse events compared to resection [1–3]. During the process of ablation, image guidance is used in planning, targeting, monitoring, and treatment response assessment [4]. Compared with CT or MR imaging, the main advantages of ultrasound guidance include better cost-effectiveness, real-time capability, and no exposure to ionizing radiation or nephrotoxicity of contrast material [5]. However, conventional two-dimensional ultrasound (2DUS) could only provide the morphological information of tumor in distinct planes and lack a direct display of tumor location in three-dimensional space, which could lead to unsatisfactory needle placement, inadequate ablative margin, and local tumor progression (LTP) [6–8]. Precise needle placement of thermal ablation is still a challenging issue and affects the prognosis of HCC patients to a substantial extent.

To enhance the accuracy of ultrasound-guided thermal ablation in HCC patients, fusion imaging (FI) technique was applied to various steps along the ablation process, such as tumor localization and ablative margin assessment [9–11]. The overlay of postoperative real-time sonograms and volume data collected preoperatively allows a direct visualization of the relationship between ablative zone and tumor, which helps the operator determine whether another overlapping ablation is needed [12]. And it has been reported that US-US FI was highly effective for ablative margin achievement and could reduce the risk of LTP after HCC thermal ablation [13,14]. The planning of ablation needle placement is highly dependent on operator’s experience, and needle placement under 2DUS guidance was more likely to be unsatisfactory and may need readjustment to the position or number of ablation needles [6]. CT or MR-US FI was used to guide needle placement when the tumor was poorly visible in ultrasound images, which improved the tumor visibility and technical feasibility of thermal ablation [15]. Positron emission tomography (PET) was used in the guidance of fluorodeoxyglucose (FDG)-avid liver tumor ablation, and ablative margin can be assessed through fusion imaging of intraoperative PET and contrast-enhanced CT [16,17]. US-US FI could also be used in the planning and assessment of needle placement, with advantages of convenience and high accuracy of registration [18,19]. However, studies of US-US FI commonly focused on the postoperative assessment of ablation zone, the value of US-US FI in preoperative planning and intraoperative guidance of ablation needle placement still needs further exploration. In a previous in vivo simulation study, it was found that three-dimensional ultrasound fusion imaging (3DUS FI) real-time guidance could improve the accuracy of ablation needle placement, especially for inexperienced operators [20]. The present study aimed to further evaluate the clinical efficacy and safety of 3DUS FI real-time guidance for precise needle placement in HCC thermal ablation.

Materials and methods

Patient selection

The study was approved by the institutional review board of our hospital (approved No. [2022] 132), and the written informed consent was waived owing to the retrospective nature of this research. All selected patients had signed informed consent and been fully informed of the necessity, adverse reactions, and complications of treatment. Clinical data of HCC patients that underwent thermal ablation in our hospital from November 2019 to October 2022 were analyzed retrospectively. The inclusion criteria of this study were: (1) confirmed HCC according to the American Association for the Study of Liver Diseases (AASLD) clinical guidelines or by biopsy; (2) single lesion with maximum diameter ≥ 1 cm and ≤ 5 cm; or multiple lesions with maximum diameter ≥ 1 cm and ≤ 3 cm, and the number ≤ 3; (3) no other therapy received before ablation, including transcatheter arterial chemoembolization (TACE), chemotherapy, radiotherapy, or targeted therapy; (4) no vascular invasion in preoperative radiological examination; (5) Child-Pugh class A/B, platelet count > 50 × 10⁹, prothrombin time ratio > 50%, and Eastern Cooperative Oncology Group (ECOG) performance status score = 0. Exclusion criteria were: (1) thermal ablation under contrast-enhanced ultrasound (CEUS) or CT/MR-US FI or 2DUS guidance; (2) invisible or blurry lesions under baseline ultrasound; (3) complete 3DUS volume data unavailable because of obstruction of ribs or lung gas; (4) presence of uncontrolled ascites; (5) patients received liver transplantation after ablation; (6) a follow-up period of less than 6 months.

Equipment and ablation procedure

3DUS FI guidance was conducted by Resona 9 ultrasound machine (Mindray, China) with a built-in navigation system (uHIT Navi). Position-sensing unit was assembled with C6-2Gs probe (1.2-6.0 MHz) through a bracket.

Radiofrequency ablation (RFA) was performed through Cool-tip RFA system (Covidien, USA) with CTRF220 radiofrequency generator (480 kHz, max output power 200 W) and 20 cm-long, 17-gauge electrodes with 2-3 cm active metallic tip. Microwave ablation (MWA) was performed through KY-2000 MWA device (Canyon, China) with 2450 MHz microwave transmitter (max output power 150 W) and 15-gauge KY-2450B microwave antenna.

All the RFA or MWA procedures were carried out by one of two experienced operators (X.Y.X and G.L.H with 13 and 8 years of ablation experience, respectively). To achieve a complete tumor ablation, the ablative margin was supposed to be at least 5 mm. If the tumor lesion was within 5 mm of surrounding important structures, such as gallbladder, diaphragm, or gastrointestinal tract, artificial ascites could be applied by infusing 5% glucose into abdominal cavity to avoid thermal injury. After ablation treatment, the needle was slowly retracted and the needle tract was cauterized to prevent bleeding and tumor implantation.

3DUS FI guidance

Firstly, 3DUS volume data of the tumor and surrounding structures were acquired preoperatively from breath-holding patients in freehand scanning mode with a slow uniform speed. After acquisition of the volume data, three-dimensional reconstruction and registration with real-time 2DUS images will be conducted automatically and immediately without changing of patients’ body position. If the accuracy of registration is unsatisfactory, a manual fine-tuning could be made by operators. (Figure 1(A))

Figure 1. 3DUS FI technique in precise needle placement for HCC thermal ablation. (A) 3DUS volume data around the tumor was acquired by freehand scanning (left). Tumor (green) was automatically segmented and a 5 mm ablative margin (red) was displayed (right). (B) The plan of needle placement was made through predetermined simulative ablation zone (light blue), and the estimated coverage rate was automatically calculated. (C) Ablation needles were placed according to the plan with 3DUS fusion imaging guidance. The planned puncture path and depth was shown in red dashed line, and the planned ablation zone was shown in red circle in real-time 2DUS. (D) Intraoperative assessment of the coverage of ablation zone was conducted through CEUS and 3DUS fusion imaging.

Secondly, cross-sections of 3DUS volume data were selected based on the largest diameter of the tumor, and the tumor could be automatically segmented and labeled in three perpendicular planes. If the automatic segmentation was unsatisfactory, manual adjustments could be made. After segmentation, the center point of the tumor was automatically calculated and an ablative margin of 5 mm could be displayed around the tumor three-dimensionally. The location and volume of tumor and ablative margin would be saved in the 3DUS volume data and projected to the real-time 2DUS image. (Figure 1(A))

Thirdly, under the guidance of 3DUS FI, operators could set virtual ablation needles with adjustable puncture path, angle, and depth. Predetermined simulative ablation zone (3 × 2 × 2 cm in RFA, 3.5 × 2.5 × 2.5 cm in MWA) would be displayed around the tip of the virtual ablation needle. With the movement of probe, the relationship of spatial position between the simulative ablation zone and the tumor (including ablative margin) can be observed in real time, and the estimated coverage rate of ablation zone would be automatically calculated to ensure a 5 mm ablative margin. The plan of needle placement could then be saved in 3DUS volume data. (Figure 1(B))

Lastly, with real-time 2DUS and 3DUS fusion imaging guidance, ablation needles were placed according to the plan. During the ablation, operators could monitor the coverage of hyperechoic ablative zone to the labeled tumor and ablative margin. After ablation according to the plan, assessment of ablative margin was conducted through intraoperative CEUS and 3DUS fusion imaging. If the non-enhancement area in CEUS did not completely cover the labeled tumor and ablative margin in fusion imaging, which means the actual ablative margin was less than 5 mm, supplementary ablation could be performed immediately. (Figure 1(C,D))

The flowchart of 3DUS FI technique was shown in Figure 2.

Figure 2. Flowchart of 3DUS FI technique.

Abbreviations: 3DUS: three-dimensional ultrasound, FI: fusion imaging, CEUS: contrast-enhanced ultrasound

Follow-up

Postoperative CEUS examination was performed within 3 days after ablation to evaluate the occurrence of early complications and the rate of primary technical success, which was defined as complete coverage of ablation zone to the tumor after first session of ablation [21]. One month after ablation, efficacy of ablation was assessed by contrast-enhanced CT/MRI (CECT/MRI), CEUS, and level of alpha-fetoprotein (AFP). After that, CEUS and AFP level were rechecked every 3 months, and CECT/MRI was rechecked every 6 months. Postoperative complications were recorded and graded according to the Society of Interventional Radiology (SIR) classification system [22].

Complete response (CR) was defined as no enhanced lesion observed in the original tumor location in CECT/MRI 1 month after ablation. Ablative margin was measured by two sophisticated doctors with extensive experience in abdominal radiology and was performed in Picture Archiving and Communication Systems (PACS) workstation according to the CECT/MRI 1 month after ablation. If insufficient ablative margin was found in postoperative CECT/MRI, the ablation zone would be monitored every 3 months through CEUS for potential recurrence. Local tumor progression (LTP) referred to newly emerged lesions at the edge of ablation foci or blood supply observed inside the ablation foci by CECT/MRI examination in CR patients during the follow-up period [21], while intrahepatic distant recurrence (IDR) and extrahepatic recurrence (ER) were defined as any newly-emerged metastatic lesion with different location of recurrence detected by CECT/MR or CEUS during the follow-up period.

Data collection and statistical analysis

The following data were prospectively collected and analyzed. (1) Patients baseline, including gender, age, basic disease (hepatitis, liver cirrhosis, and portal hypertension), Child-Pugh class, and level of AFP. (2) Basic information of tumors and treatments, including tumor type (primary or recurrent), location, size, adjacent structure (with minimum distance < 5 mm), treatment type (RFA or MWA), and whether artificial ascites was used. (3) Operation-related parameters, including operation time, rate of successful registration, numbers of ablation needles, ablation points, puncture times, and rate of intraoperative supplementary ablation. (4) Follow-up data, including postoperative complications, rate of primary technical success, rate of adequate ablative margin, CR, LTP, IDR, ER, disease-free survival (DFS), and overall survival (OS).

Continuous variables were expressed as mean ± SD or median (P25-P75) according to the distribution, and categorical variables were expressed as number of cases (percentage). Kaplan-Meier method was used to calculate the rate of LTP, DFS, and OS among patients. For comparative analysis between patients treated with RFA and MWA, Chi-square test or Fisher’s exact test was used for complication rate and adequate ablative margin achievement rate, and Log-rank test was used for rates of LTP, IDR, ER, DFS, and OS. p < 0.05 was considered statistical difference in comparisons. All statistical analysis was conducted using SPSS 22.0 (SPSS Inc., Chicago, USA).

Results

General information

A total of 57 patients with 60 lesions with 3DUS FI-guided ablation were included in this research. The average age of included patients was 57.3 ± 12.1 years old. Thirty patients were diagnosed with primary HCC and 27 patients had recurrent HCC. The average diameter of ablated tumors was 2.2 ± 0.6 cm. The adjacent anatomical structures were judged by preoperative CT/MRI. A total of 14 tumors (23%) were close to major blood vessels (diameter > 3 mm), 13 tumors (22%) were close to diaphragm, and 2 tumors (3%) were close to gastrointestinal tract. The detailed information of other baseline characteristics was listed in Table 1.

Table 1. Baseline characteristics of patients.

Characteristics	Value
Number of patients	57
Gender (Male/Female)	47/10
Age (year)	57.3 ± 12.1
Etiological cause of HCC
Hepatitis B	55 (96.4)
Hepatitis C	1 (1.8)
Non-hepatitis	1 (1.8)
Type of HCC (primary/recurrent)	30/27
Number of tumors in each patient (1/≥2)	49/8
Liver cirrhosis (Yes/No)	21/36
Portal hypertension (Yes/No)	9/48
AFP (≤20 ng/mL/>20 ng/mL)	33/24
Child-Pugh grade
A	53 (93.0)
B	4 (7.0)
Number of tumors	60
Tumor size (cm)	2.2 ± 0.6
Proximity to critical structures
Major blood vessels	14 (23.3)
Diaphragm	13 (21.7)
Gastrointestinal tract	2 (3.3)

Abbreviations: HCC hepatocellular carcinoma, AFP alpha-fetoprotein.

Operation-related parameters

RFA and MWA were performed in 42 lesions (70%) and 18 lesions (30%) respectively, and artificial ascites were used in eight lesions (13%). For 3DUS FI guidance process, a successful registration was defined as the deviation distance less than 2 mm between fused images within 3 times of registration attempts, and the rate of successful registration was 100% (60/60). The median of 3DUS FI-guided operation time was 16 (12-22) minutes, with the median of ablation needles being 2.0 (1.0-2.0). The median of ablation points was 2.0 (2.0-3.0), and the median of puncture times was 2.0 (2.0-2.0). During the operation, 10 lesions (17%) received supplementary ablation because of inadequate ablative margin detected by intraoperative CEUS. The specific distribution of the number of ablation needles, puncture times, and ablation points among all patients was provided in Table 2.

Table 2. Detailed operation-related parameters.

Parameters	Value
Number of ablation needles
1	16 (26.7)
2	42 (70.0)
3	2 (3.3)
Number of puncture times
1	12 (20.0)
2	34 (56.7)
3	5 (8.3)
4	8 (13.3)
6	1 (1.7)
Number of ablation points
1	9 (15.0)
2	34 (56.7)
3	7 (11.6)
4	8 (13.3)
6	1 (1.7)
8	1 (1.7)

Effect of ablation treatment

The rate of primary technical success was 100% (60/60), and none of the patients needed supplementary operation according to the postoperative ultrasonography performed within 3 days after ablation. No major complications or ablation-related deaths were observed in any patients. Minor complications were observed in 8 patients (14%). Three patients (5%) had self-limited hepatic hemorrhage, two patients (4%) had postoperative fever, two patients (4%) had pleural effusion, and one patient (2%) had both fever and pleural effusion. No statistical difference in the rate of complications was found between patients treated with RFA and MWA (15% and 12%, respectively, p = 1.000). According to the CECT/MRI examination 1 month after ablation, the CR rate was 100% (60/60) (Figures 3 and 4). Among these patients, 35 patients with 36 lesions took the CECT/MRI examination in our hospital, and 30 lesions (83%) achieved adequate ablative margin (> 5 mm), while 6 lesions (17%) had inadequate ablative margin (< 5 mm). Between RFA group and MWA group, the rate of adequate ablative margin achievement (82% and 89% respectively) has no statistical difference (p = 1.000).

Figure 3. A 37-year-old female with hepatocellular carcinoma in segment VIII of liver. (A) Preoperative MRI scan showed one HCC lesion in segment VIII with the diameter measured to be 2.3 cm. (B) On baseline ultrasound, the lesion appeared hypoechoic with clear boundary and regular morphology. (C) After acquisition of 3DUS volume data and reconstruction, the tumor was automatically segmented and labeled in green color, and an ablative margin of 5 mm was displayed around the tumor in purple color. (D) Two ablation needles were planned to be placed, and the simulative ablation zone had covered the tumor and ablative margin completely through assessment in different plane. (E) Puncture of ablation needles according to the plan. The tips of ablation needles were shown by red arrows. (F) When ablation is done, the accuracy of registration of real-time 2DUS image (left) and 3DUS reconstructive image (right) were confirmed again. The tumor and ablative margin were completely covered by the hyperechoic ablation zone. (G) MRI at one month after ablation showed complete ablation of the tumor with ablative margin greater than 5 mm.

Figure 4. A 58-year-old male with hepatocellular carcinoma in segment VI of liver. (A) Preoperative MRI showed one HCC lesion in segment VI with diameter measured to be 2.1 cm. (B) In baseline ultrasound, the lesion appeared hypoechoic. (C) Tumor (in green color) and ablative margin (in purple color) were labeled in 3DUS reconstructive image (right) and projected to real-time 2DUS image (left). (D) With adjustment of the puncture path, angle, and depth, two needles were planned to be placed to ensured complete ablation. (E) During the needle placement process, the labels of lesion and ablative margin could be hidden in real-time 2DUS image if needed. The red arrows showed the tip of two ablation needles. (F) The tumor and ablative margin were completely covered by the hyperechoic ablation zone (left), and the scanning plane could be displayed in three-dimensional way to assist comprehensive assessment (middle and right). (G) MRI at one month after ablation showed complete ablation of the tumor with ablative margin greater than 5 mm.

The median follow-up time was 24.5 months (17.5-31.4 months). During the follow-up period, LTP occurred in 5 lesions, with the LTP rate to be 8%. Among these lesions, 2 and 3 lesions were close to major blood vessels and diaphragm respectively, and the ablative margin was assessed in 3 lesions in our hospital, which were all less than 5 mm. The 1- and 2-year LTP rates calculated by Kaplan-Meier method were 7.0% and 9.4% respectively, and no statistical difference was found between RFA group and MWA group (p = 0.584). IDR occurred in 15 patients (26%), and ER occurred in 10 patients (18%). During the follow-up period, 3 patients died of multiple extrahepatic metastases 12, 16, and 17 months after ablation respectively. The 1- and 2-year DFS rates were 76.1% and 65.6%. The 1- and 2-year OS rates were 98.1% and 94.0%. Between RFA group and MWA group, there is no statistical difference in rates of IDR (p = 0.100), ER (p = 0.481), DFS (p = 0.333), or OS (p = 0.258).

Discussion

3DUS FI guidance for thermal ablation combines the technologies of US-US FI, three-dimensional reconstruction and visualization, which is helpful for tumor localization and ablative margin assessment with advantages of real-time capability, accuracy of registration, and convenience of operation [23,24]. In our previous in vivo simulation study, it was revealed that 3DUS FI guidance could enhance the accuracy of ablation needle placement and was conducive to the achievement of adequate ablative margin [20]. In this study, we further explore the clinical value of precise needle placement through 3DUS FI guidance in HCC thermal ablation and demonstrate the efficacy and safety.

The technique of fusion imaging has substantially promoted the development of thermal ablation in HCC treatment, and the clinical application is affected by the convenience and reliability of technique to a great extent. CT/MR-US FI has been used in the location of inconspicuous lesions and assessment of ablative margin [15,25]. However, in comparison with multi-modality fusion imaging, the rate of successful registration in US-US FI was reported to be higher, which was possibly because of the anatomical consistency in mono-modality imaging [19]. Similarly, the convenience and operability of US-US FI were reflected in this research with the successful registration rate to be 100%.

The application of US-US FI in intraoperative evaluation of ablation area could assist the operator in determining the necessity of supplementary ablation. Xv et al. reported a supplementary ablation rate of 34% under 3DUS-CEUS FI [14], and the corresponding supplementary ablation rate in our study was only 17%. Besides, the CR rate achieved 100% under 3DUS FI guidance in our study, compared with the CR rate of 86–97% under 2DUS or CEUS guidance [26–28] and 80–96% under CT or MR guidance [29,30], which suggested the preciseness of needle placement plan-making and real-time guidance based on 3DUS FI technique. For those patients who had preoperative needle placement planning but still received supplementary ablations, a possible reason is the deviation between actual and theoretical range of ablation zone. The actual ablation range could be influenced by ablation power, ablation time, condition of liver cirrhosis, and tumor location [31]. In further research, a simulative ablation zone that can be adjusted based on patient and operation condition would be helpful to better estimate the coverage to tumor and ablative margin.

Ablative margin is one of the most important factors affecting the LTP rate and prognosis in HCC patients. It was reported that an ablative margin < 5 mm was significantly correlated with the occurrence of LTP [8,32], and was associated with poorer survival in patients undergoing thermal ablation [33]. In previous studies, the rate of adequate ablative margin was significantly higher with US-US FI guidance (88%-89%) than with 2DUS guidance (44–47%) [13,34]. A similar trend was also observed in this study with the rate of adequate ablative margin to be 83%. It was observed that a few patients achieved adequate ablative margins during the intraoperative evaluation, but the postoperative CT/MR showed inadequate ablative margins. The possible reasons are as follows. Firstly, the respiratory motion of patients during operation may result in registration deviation of fusion images, which compromises the accuracy of intraoperative evaluation and could be corrected by respiratory compensation in further study. Secondly, the liver may undergo deformation after ablation, and the clarity of the ablation zone may be reduced due to gas interference during intraoperative CEUS, which could also affect the evaluation accuracy.

A feature of the ablated lesions included in this study was the relatively high proportion of proximity to critical structures. A total of 28 lesions (47%) were close to major blood vessels, diaphragm, or gastrointestinal tract with a linear distance less than 5 mm, which included all 5 lesions with LTP. The proximity to critical structures would increase the risk of puncture and ablation, and limit the size of ablation area, which was a possible reason for a higher LTP rate in these lesions. Additionally, it was reported that microvascular invasion (MVI) could be detected in around 30-40% of HCC patients who underwent hepatic resection [35,36], and among 5–12% of these patients, the distance from MVI to tumor was greater than 1 cm [35,37]. Therefore, appropriately expanding the ablative margin in patients at higher risk of MVI might reduce the incidence of LTP.

The accuracy of needle placement is important in successful ablation treatment. The advantages of 3DUS FI guidance can be summarized as follows. Firstly, the three-dimensional information of tumor and surrounding structures can be collected and provided to operators, so it would be easier to make and adjust the placement plan of ablation needle. Secondly, the real-time display of ablation plan and actual needle puncture enables the operator to get fast feedback and make needle placement more accurate. Thirdly, 3DUS FI guidance technique can integrate preoperative planning, intraoperative guidance, and postoperative evaluation, thus making thermal ablation of HCC a one-stop treatment.

In recent years, the FI technique has been increasingly applied in HCC ablation with advantages in localization of tumor and assessment of ablation zone. A recent meta-analysis of FI-guided ablation showed lower risk of LTP and lower incidence of overall complications compared to conventional 2DUS- or CEUS-guided ablation [38]. In the analysis of patients with different tumor characteristics, it was reported that the FI technique could be more effective in tumors with larger size or risky location [39,40]. Besides, the safety of ablation could be improved by FI technique when the tumor was located close to major bile duct [41,42]. In future research, focusing on liver tumors with larger size or high-risk location may better demonstrate the value of FI technique.

The main limitations of this study were the relatively small sample size and lack of comparison to other guidance techniques. The single-center retrospective study design was another limitation, which would lead to inevitable bias in patient selection. Further prospective studies with larger sample size could have better analysis of treatment response and prognosis of patients. Besides, the average size of tumors included in this study was relatively small, and the advantages of 3DUS FI in guiding precise needle placement might be more pronounced in the ablation of larger tumors.

In conclusion, three-dimensional ultrasound fusion imaging technique may improve the needle placement of thermal ablation for HCC and reduce the rate of LTP.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data and material will be made available on request.

Additional information

Funding

This work was supported by the Major Research Plan of the National Natural Science Foundation of China [grant number 92059201], and Guangzhou Science and Technology Program [grant number 202201011157].