Leukemia is a common, multiple and dangerous blood disease, whose early diagnosis and treatment are very important. At present, the diagnosis of leukemia heavily relies on morphological examination of blood cell images by pathologists, which is tedious and time-consuming. Meanwhile, the diagnostic results are highly subjective, which may lead to misdiagnosis and missed diagnosis. To address the gap above, we proposed an improved Vision Transformer model for blood cell recognition. First, a faster R-CNN network was used to locate and extract individual blood cell slices from original images. Then, we split the single-cell image into multiple image patches and put them into the encoder layer for feature extraction. Based on the self-attention mechanism of the Transformer, we proposed a sparse attention module which could focus on the discriminative parts of blood cell images and improve the fine-grained feature representation ability of the model. Finally, a contrastive loss function was adopted to further increase the inter-class difference and intra-class consistency of the extracted features. Experimental results showed that the proposed module outperformed the other approaches and significantly improved the accuracy to 91.96% on the Munich single-cell morphological dataset of leukocytes, which is expected to provide a reference for physicians’ clinical diagnosis.
ObjectiveTo summarize the experience of robot-assisted lung basal segmentectomy, and analyze the clinical application value of intersegmental tunneling and pulmonary ligament approach for S9 and/or S10 segmentectomy. MethodsThe clinical data of 78 patients who underwent robotic lung basal segmentectomy in our hospital between January 2020 to May 2022 were retrospectively reviewed. There were 32 males and 46 females with a median age of 50 (33-72) years. The patients who underwent S9 and/or S10 segmentectomy were divided into a single-direction group (pulmonary ligament approach, n=19) and a bi-direction group (intersegmental tunneling, n=19) according to different approaches, and the perioperative outcomes between the two groups were compared. ResultsAll patients successfully completed the operation, without conversion to thoracotomy and lobectomy, serious complications, or perioperative death. The median operation time was 100 (40-185) min, the blood loss was 50 (10-210) mL, and the median number of dissected lymph nodes was 3 (1-14). There were 4 (5.1%) patients with postoperative air leakage, and 4 (5.1%) patients with hydropneumothorax. No patient showed localized atelectasis or lung congestion at 6 months after the operation. Further analysis showed that there was no significant difference in the operation time, blood loss, thoracic drainage time, complications or postoperative hospital stay between the single-direction and bi-direction groups (P>0.05). However, the number of dissected lymph nodes of the bi-direction group was more than that of the single-direction group [6 (1-13) vs. 5 (1-9), P=0.040]. ConclusionThe robotic lung basal segmentectomy for pulmonary nodules is safe and effective. The perioperative results of robotic S9 and/or S10 complex segmentectomy using intersegmental tunneling and pulmonary ligament approach are similar.