Failure to identify difficult intubation is the leading cause of anesthesia-related death and morbidity. Despite preoperative airway assessment, 75-93% of difficult intubations are unanticipated, and airway examination methods underperform, with sensitivities of 20-62% and specificities of 82-97%. To overcome these impediments, we aim to develop a deep learning model to identify difficult to intubate patients using frontal face images. We proposed an ensemble of convolutional neural networks which leverages a database of celebrity facial images to learn robust features of multiple face regions. This ensemble extracts features from patient images (n = 152) which are subsequently classified by a respective ensemble of attention-based multiple instance learning models. Through majority voting, a patient is classified as difficult or easy to intubate. Whereas two conventional bedside tests resulted in AUCs of 0.6042 and 0.4661, the proposed method resulted in an AUC of 0.7105 using a cohort of 76 difficult and 76 easy to intubate patients. Generic features yielded AUCs of 0.4654-0.6278. The proposed model can operate at high sensitivity and low specificity (0.9079 and 0.4474) or low sensitivity and high specificity (0.3684 and 0.9605). The proposed ensembled model outperforms conventional bedside tests and generic features. Side facial images may improve the performance of the proposed model. The proposed method significantly surpasses conventional bedside tests and deep learning methods. We expect our model will play an important role in developing deep learning methods where frontal face features play an important role.

More humans have died of tuberculosis (TB) than any other infectious disease and millions still die each year. Experts advocate for blood-based, serum protein biomarkers to help diagnose TB, which afflicts millions of people in high-burden countries. However, the protein biomarker pipeline is small. Here, we used the Diversity Outbred (DO) mouse population to address this gap, identifying five protein biomarker candidates. One protein biomarker, serum CXCL1, met the World Health Organization’s Targeted Product Profile for a triage test to diagnose active TB from latent M.tb infection (LTBI), non-TB lung disease, and normal sera in HIV-negative, adults from South Africa and Vietnam. To find the biomarker candidates, we quantified seven immune cytokines and four inflammatory proteins corresponding to highly expressed genes unique to progressor DO mice. Next, we applied statistical and machine learning methods to the data, i.e., 11 proteins in lungs from 453 infected and 29 non-infected mice. After searching all combinations of five algorithms and 239 protein subsets, validating, and testing the findings on independent data, two combinations accurately diagnosed progressor DO mice: Logistic Regression using MMP8; and Gradient Tree Boosting using a panel of 4: CXCL1, CXCL2, TNF, IL-10. Of those five protein biomarker candidates, two (MMP8 and CXCL1) were crucial for classifying DO mice; were above the limit of detection in most human serum samples; and had not been widely assessed for diagnostic performance in humans before. In patient sera, CXCL1 exceeded the triage diagnostic test criteria (>90% sensitivity; >70% specificity), while MMP8 did not. Using Area Under the Curve analyses, CXCL1 averaged 94.5% sensitivity and 88.8% specificity for active pulmonary TB (ATB) vs LTBI; 90.9% sensitivity and 71.4% specificity for ATB vs non-TB; and 100.0% sensitivity and 98.4% specificity for ATB vs normal sera. Our findings overall show that the DO mouse population can discover diagnostic-quality, serum protein biomarkers of human TB.

Recent methods in computational pathology have trended towards semi- and weakly-supervised methods requiring only slide-level labels. Yet, even slide-level labels may be absent or irrelevant to the application of interest, such as in clinical trials. Hence, we present a fully unsupervised method to learn meaningful, compact representations of WSIs. Our method initially trains a tile-wise encoder using SimCLR, from which subsets of tile-wise embeddings are extracted and fused via an attention-based multiple-instance learning framework to yield slide-level representations. The resulting set of intra-slide-level and inter-slide-level embeddings are attracted and repelled via contrastive loss, respectively. This resulted in slide-level representations with self-supervision. We applied our method to two tasks— (1) non-small cell lung cancer subtyping (NSCLC) as a classification prototype and (2) breast cancer proliferation scoring (TUPAC16) as a regression prototype—and achieved an AUC of 0.8641 ± 0.0115 and correlation (R²) of 0.5740 ± 0.0970, respectively. Ablation experiments demonstrate that the resulting unsupervised slide-level feature space can be fine-tuned with small datasets for both tasks. Overall, our method approaches computational pathology in a novel manner, where meaningful features can be learned from whole-slide images without the need for annotations of slide-level labels. The proposed method stands to benefit computational pathology, as it theoretically enables researchers to benefit from completely unlabeled whole-slide images.

Background: Identification of bladder layers is a necessary prerequisite to bladder cancer diagnosis and prognosis. We present a method of multi-class image segmentation, which recognizes urothelium, lamina propria, muscularis propria, and muscularis mucosa layers as well as regions of red blood cells, cauterized tissue, and inflamed tissue from images of hematoxylin and eosin stained slides of bladder biopsies. Methods: Segmentation is carried out using a U-Net architecture. The number of layers was either, eight, ten, or twelve and combined with a weight initializers of He uniform, He normal, Glorot uniform, and Glorot normal. The most optimal of these parameters was found by through a seven-fold training, validation, and testing of a dataset of 39 whole slide images of T1 bladder biopsies. Results: The most optimal model was a twelve layer U-net using He normal initializer. Initial visual evaluation by an experienced pathologist on an independent set of 15 slides segmented by our method yielded an average score of 8.93 ± 0.6 out of 10 for segmentation accuracy. It took only 23 min for the pathologist to review 15 slides (1.53 min/slide) with the computer annotations. To assess the generalizability of the proposed model, we acquired an additional independent set of 53 whole slide images and segmented them using our method. Visual examination by a different experienced pathologist yielded an average score of 8.87 ± 0.63 out of 10 for segmentation accuracy. Conclusions: Our preliminary findings suggest that predictions of our model can minimize the time needed by pathologists to annotate slides. Moreover, the method has the potential to identify the bladder layers accurately. Further development can assist the pathologist with the diagnosis of T1 bladder cancer.

Because most humans resist Mycobacterium tuberculosis infection, there is a paucity of lung samples to study. To address this gap, we infected Diversity Outbred mice with M. tuberculosis and studied the lungs of mice in different disease states. After a low-dose aerosol infection, progressors succumbed to acute, inflammatory lung disease within 60 days, while controllers maintained asymptomatic infection for at least 60 days, and then developed chronic pulmonary tuberculosis (TB) lasting months to more than 1 year. Here, we identified features of asymptomatic M. tuberculosis infection by applying computational and statistical approaches to multimodal data sets. Cytokines and anti- M . tuberculosis cell wall antibodies discriminated progressors vs controllers with chronic pulmonary TB but could not classify mice with asymptomatic infection. However, a novel deep-learning neural network trained on lung granuloma images was able to accurately classify asymptomatically infected lungs vs acute pulmonary TB in progressors vs chronic pulmonary TB in controllers, and discrimination was based on perivascular and peribronchiolar lymphocytes. Because the discriminatory lesion was rich in lymphocytes and CD4 T cell-mediated immunity is required for resistance, we expected CD4 T-cell genes would be elevated in asymptomatic infection. However, the significantly different, highly expressed genes were from B-cell pathways (e.g., Bank1 , Cd19 , Cd79 , Fcmr , Ms4a1 , Pax5 , and H2-Ob ), and CD20+ B cells were enriched in the perivascular and peribronchiolar regions of mice with asymptomatic M. tuberculosis infection. Together, these results indicate that genetically controlled B-cell responses are important for establishing asymptomatic M. tuberculosis lung infection.

Identification of bladder layers from tissue biopsies is the first step towards an accurate diagnosis and prognosis of bladder cancer. We present an automated Bladder Image Analysis System (BLIAS) that can recognize urothelium, lamina propria, and muscularis propria from images of H and E-stained slides of bladder biopsies. Furthermore, we present its clinical application to automate risk stratification of T1 bladder cancer patients based on the depth of lamina propria invasion. The method uses multidimensional scaling and transfer learning in conjunction with convolutional neural networks to identify different bladder layers from H and E images of bladder biopsies. The method was trained and tested on eighty whole slide images of bladder cancer biopsies. Our preliminary findings suggest that the proposed method has good agreement with the pathologist in identification of different bladder layers. Additionally, given a set of tumor nuclei within lamina propria, it has the potential to risk stratify T1 bladder cancer by computing the distance from this set to urothelium and muscularis propria. Our results suggest that a pretrained network trained via transfer learning is better in identifying bladder layers than a conventional deep learning paradigm.