Efficiency and impact factors of anatomical intelligence for breast and hand‐held ultrasound in lesion detection

To investigate the efficiency and impact factors of anatomical intelligence for breast (AI‐Breast) and hand‐held ultrasound (HHUS) in lesion detection.

HHUS was performed by breast imaging radiologists (Group A) or general radiologists (Group B). The radiologists in Group A had 8-10 years of experience in breast ultrasound (X.X and Y.J) while those in Group B had a 5-to 8-year expertise in ultrasound diagnosis (L.F and J.J).

| AI-Breast ultrasound
A color ultrasound instrument (Philips EPIQ5) equipped with a variable-frequency probe (eL18-4) was used. This technology comprised a frequency range of 2-22 MHz, a receivable magnetic field signal, and a AI-Breast magnetic positioning device mattress.
In Group AI, the technician conducted the scanning as follows: the patient lying on the special mattress took a supine position with arms raised above the head, and the targeted breast was raised by appropriate lateral rotation. Triangular support behind the examined side helped fix the position and place the breast into an effective magnetic field. Before scanning, the scan area was determined through locating the nipple and the lower, inner, and outer margin of the breast ( Figure 1A). When scanning the breast, the transducer on the screen display moves with the scanning hand Subsequently, the breast data were saved in video by the technician, imported into a computer, and interpreted using professional software by the radiologists. The time of interpretation, and lesion number and location were recorded.

| HHUS examination
The previously described Philips EPIQ5 instrument equipped with the eL18-4 probe was employed. Examinations in Groups A and B began with patients taking a supine position and fully exposing their chest. A comprehensive breast scanning was performed radially or alternating up and down, left and right. During the examination, the images were saved and the number and specific location of the lesions were recorded.

| Number and categorization of breast lesions
Breast lesion detection and patient clinical data were analyzed in the three groups. If the number of lesions was the same in the three groups, there was no controversy. If the number of lesions was different, two radiologists in Group A reviewed the AI-Breast image information and saved 2D ultrasound images to make a judgment. If necessary, patients were called for reexamination.
Criteria for benign and malignant breast lesions: (1) lesions with BI-RADS categories 4 and above were confirmed as benign or malignant by surgery or puncture; (2) lesions with BI-RADS 3 and below were confirmed by surgery. In the cases where neither surgery nor puncture was conducted, all lesions were followed up for at least 2 years (≥2 times). Ultrasound images of the last 2 years were reviewed. No significant change in the size and ultrasound characteristics of the lesions was judged benign. Lastly, 10% of patients were randomly selected for MRI to exclude false positives.

| Statistical analysis
The MedCal software was used for statistical analyses. Measurement data were expressed as mean ± SD. Differences in measurement data between two groups were analyzed by the Mann-Whitney test, with P < 0.05 determining statistical significance.
Differences in measurement data between multiple groups were analyzed by the ANOVA test, with P < 0.05 indicating a significant difference between at least two groups.   (Table 1). Furthermore, scan times significantly increased with breast size in the three groups. In particular, T AI scan had a positive linear correlation with breast size (r AI scan = 0.745) (Tables 2 and 4). On the other hand, the scan times of benign cases from Group A and group B were significantly longer than those of malignant cases (P < 0.05). T AI scan did not correlate with benign or malignant lesions, and there were no significant differences (P = 0.688) (Tables 2 and 4).
No statistically significant difference was noted between Group AI and Group A (P = 0.367). However, differences were found when comparing Group A and Group AI, respectively, with Group B (both P < 0.05).
Detection rates in Group A and Group AI did not correlate with number of lesions, and there was no statistical significance. For its part, the detection rate in Group B had a negative linear correlation with the number of lesions (r b = À0.436, P<0.001). In Group B, patients with number of lesions ≥2 and C cup and above had a decreased detection rate that was significantly lower than those in T A B L E 1 Basic characteristics of enrolled patients. Group A and Group AI (P < 0.05) (Tables 3 and 4). Moreover, detection rates throughout the three groups did not correlate with breast size (P > 0.05) (Tables 3 and 4).
For benign lesions, missed diagnosis rates in Group A, Group AI, and Group B were 7%, 8%, and 20%, respectively. Group B had higher rates than Group A and Group AI (both P < 0.05) while there were no statistical differences between Group AI and Group A (P = 0.629). For malignant lesions, missed diagnosis rates in Group A, Group AI, and Group B were 4%, 8%, and 14%, respectively. There were no statistical differences among them (P > 0.05) ( Table 5).
The possible causes for missing breast lesions on AI-Breast and HHUS are shown in

| DISCUSSION
Mammography is considered the main imaging method for the reduction of breast cancer mortality. However, the sensitivity of this method is lower in dense breasts. 11,12 In China and other Asian countries, the proportion of women with dense breasts are higher than in western countries. Ultrasound systems overcome this issue with highscreening efficiency in dense breasts. [13][14][15][16] Because of this, the ultra- T A B L E 3 Impact of number of lesions and breast size on the detection rates in the three groups.  It has been reported that the lesion detection rate of HHUS may correlate with breast size, lesion size, and number of lesions. [26][27][28][29][30] The smaller the breast lesion is, the lower the detection rate. When the breast size is large, the lesions tend to locate in the deep tissue, which increases the difficulty in the ultrasound analysis and reduces the detection rate. In this study, the detection rate of the AI-breast ultra-    radiologists was the shortest, while that of AI-Breast was relatively shorter than general radiologists. Nevertheless, due to the different characteristics of scan and diagnosis in AI-Breast ultrasound, extra interpretation time was needed; therefore, the whole examination time was longer than those of the other two groups.

HHUS
This study has some limitations as a single-center project with data from a hospital. The radiologists had relatively wide experience in breast diagnosis; thus the AI-Breast ultrasound had a high detection rate of breast lesions. Multicenter and large-sample databases are needed to assess the efficiency of the AI-Breast ultrasound in asymptomatic patients and the primary hospital screening. Indeed, electromagnetic tracking devices have their limitations. An electromagnetic generator is usually placed near the head of the scan bed, and the magnetic sensor has a limited detection range. Thus, the patient needs to be close to the generator as much as possible.
Simultaneously, patients need to remove metal and electronic objects from their bodies before the scan to prevent interference in the magnetic field.

| CONCLUSIONS
Above all, AI-Breast ultrasound is similar to the conventional HHUS but achieves a separation between diagnosis and scan. Image scan is standardized, reproducible, and provides more objective location information in the daily clinical practice. 10 This grants a research basis for the remote consultation and breast cancer detection. Its advantages, such as improving operator dependency, increasing the detection sensitivity, technician training with less time and cost, and focusing breast diagnosis to radiologists, have better social and economic benefits. AI-Breast ultrasound may be a potential approach for breast cancer detection in the primary hospitals.