Efficacy and Safety of Preoperative vs. Intraoperative Computed Tomography-Guided Lung Tumor Localization: A Randomized Controlled Trial

Background: Thoracoscopic removal of small pulmonary nodules is traditionally accomplished through a two-step approach—with lesion localization in a CT suite as the first step followed by lesion removal in an operating room as the second step. While the advent of hybrid operating rooms (HORs) has fostered our ability to offer a more patient-tailored approach that allows simultaneous localization and removal of small pulmonary nodules within a single-step, randomized controlled trials (RCTs) that compared the two techniques (two- vs. single-step) are still lacking.Methods: This is a RCT conducted in an academic hospital in Taiwan between October 2018 and December 2019. To compare the outcomes of traditional two-step preoperative CT-guided small pulmonary nodule localization followed by lesion removal vs. single-step intraoperative CT-guided lesion localization with simultaneous removal performed by a dedicated team of thoracic surgeons. The analysis was conducted in an intention-to-treat fashion. The primary study endpoint was the time required for lesion localization. Secondary endpoints included radiation doses, other procedural time indices, and complication rates.Results: A total of 24 and 25 patients who received the single- and two-step approach, respectively, were included in the final analysis. The time required for lesion localization was significantly shorter for patients who underwent the single-step procedure (median: 13 min) compared with the two step-procedure (median: 32 min, p < 0.001). Similarly, the radiation dose was significantly lower for the former than the latter (median: 5.64 vs. 10.65 mSv, respectively, p = 0.001).Conclusions: The single-step procedure performed in a hybrid operating room resulted in a simultaneous reduction of both localization procedural time and radiation exposure.

Deep metric attention learning for skin lesion classification in dermoscopy images

Currently, convolutional neural networks (CNNs) have made remarkable achievements in skin lesion classification because of their end-to-end feature representation abilities. However, precise skin lesion classification is still challenging because of the following three issues: (1) insufficient training samples, (2) inter-class similarities and intra-class variations, and (3) lack of the ability to focus on discriminative skin lesion parts. To address these issues, we propose a deep metric attention learning CNN (DeMAL-CNN) for skin lesion classification. In DeMAL-CNN, a triplet-based network (TPN) is first designed based on deep metric learning, which consists of three weight-shared embedding extraction networks. TPN adopts a triplet of samples as input and uses the triplet loss to optimize the embeddings, which can not only increase the number of training samples, but also learn the embeddings robust to inter-class similarities and intra-class variations. In addition, a mixed attention mechanism considering both the spatial-wise and channel-wise attention information is designed and integrated into the construction of each embedding extraction network, which can further strengthen the skin lesion localization ability of DeMAL-CNN. After extracting the embeddings, three weight-shared classification layers are used to generate the final predictions. In the training procedure, we combine the triplet loss with the classification loss as a hybrid loss to train DeMAL-CNN. We compare DeMAL-CNN with the baseline method, attention methods, advanced challenge methods, and state-of-the-art skin lesion classification methods on the ISIC 2016 and ISIC 2017 datasets, and test its generalization ability on the PH2 dataset. The results demonstrate its effectiveness.

Preoperative Localization in Colonic Surgery (PLoCoS Study): a multicentric experience on behalf of the Italian Society of Colorectal Surgery (SICCR)

The aim of this prospective multicentric study was to compare the accurate colonic lesion localization ratio between CT and colonoscopy in comparison with surgery. All consecutive patients from 1st January to 31st December 2019 with a histologically confirmed diagnosis of dysplastic adenoma or adenocarcinoma with planned elective, curative colonic resection who underwent both colonoscopy and CT scans were included. Each patient underwent conventional colonoscopy and CT to stage the tumour, and the localization results of each procedure were registered. CT and colonoscopic localization were compared with surgical localization, adopted as the reference. Our analysis included 745 patients from 23 centres. After comparing the accuracy of colonoscopy and CT (for visible lesions) in localizing colonic lesions, no significant differences were found between the two preoperative tools (510/661 vs 499/661 correctly localized lesions, p = 0.518). Furthermore, after analysing only the patients who underwent complete colonoscopy and had a visible lesion on CT, no significant difference was observed between conventional colonoscopy and CT (331/427 vs 340/427, p = 0.505). Considering the intraoperative localization results as a reference, a comparison between colonoscopy and CT showed that colonoscopy significantly failed to correctly locate the lesions localized in the descending colon (17/32 vs 26/32, p = 0.031). We did not identify an advantage in using CT to localize colonic tumours. In this setting, colonoscopy should be considered the reference to properly localize lesions; however, to better identify lesions in the descending colon, CT could be considered a valuable tool to improve the accuracy of lesion localization.

Deep Learning-Based Three-Dimensional Oral Conical Beam Computed Tomography for Diagnosis

In order to deeply study oral three-dimensional cone beam computed tomography (CBCT), the diagnosis of oral and facial surgical diseases based on deep learning was studied. The utility model related to a deep learning-based classification algorithm for oral neck and facial surgery diseases (deep diagnosis of oral and maxillofacial diseases, referred to as DDOM) is brought out; in this method, the DDOM algorithm proposed for patient classification, lesion segmentation, and tooth segmentation, respectively, can effectively process the three-dimensional oral CBCT data of patients and carry out patient-level classification. The segmentation results show that the proposed segmentation method can effectively segment the independent teeth in CBCT images, and the vertical magnification error of tooth CBCT images is clear. The average magnification rate was 7.4%. By correcting the equation of R value and CBCT image vertical magnification rate, the magnification error of tooth image length could be reduced from 7.4. According to the CBCT image length of teeth, the distance R from tooth center to FOV center, and the vertical magnification of CBCT image, the data closer to the real tooth size can be obtained, in which the magnification error is reduced to 1.0%. Therefore, it is proved that the 3D oral cone beam electronic computer based on deep learning can effectively assist doctors in three aspects: patient diagnosis, lesion localization, and surgical planning.

Advantages and Challenges of Using Breast Biopsy Markers

Percutaneous image-guided biopsy procedures are the standard of care for histologic assessment of suspicious breast lesions. Post-biopsy tissue markers (clips) optimize patient management by allowing for assessment on follow-up imaging and precise lesion localization. Markers are used to ensure accurate correlation between imaging modalities, guide preoperative localization for malignant and high-risk lesions, and facilitate accurate identification of benign lesions at follow-up. Local practices differ widely, and there are no data detailing the exact frequency of use of clips for different breast biopsies. There are many indications for biopsy marker deployment, and some difficulties may be encountered after placement. The placement of biopsy markers has many advantages and few disadvantages, such that deployment should be routinely used after percutaneous biopsy procedures with rare exception.

Intracerebral Hemorrhage, Visual Hallucination and COVID-19: What Is the Connection? A Case-Related Review of the Literature on Peduncular Hallucinosis Following Intracerebral Hemorrhage

Previous publications have discussed the occurrence of intracerebral hemorrhages, hallucinations and psychosis in COVID-19 patients. In this article, we have reviewed the literature on the subject while depicting the case of a 63-year-old female patient who suffered from an intracerebral hemorrhage in the right basal ganglia and thalamus two weeks after a COVID-19 diagnosis and who developed a visual hallucinosis shortly after. We concluded that, while there may be a correlation between COVID-19 and hallucinations according to current literature, more research is yet needed to clarify. In our case, we rather interpreted the hallucinations in the context of a peduncular hallucinosis related to the intracerebral hemorrhage. We compared our patient’s lesion localization to other 15 reported cases of peduncular hallucinations following intracerebral hemorrhages reported on Pubmed. In summary, the lesions were localized in the pons in 52.9% of the cases, 17.7% were in the thalamus and/or the basal ganglia, 17.7% in the mesencephalon and respectively 5.8% in the temporal and occipital lobe. The distribution pattern we found is consistent with the previously proposed mechanism behind peduncular hallucinations.