Unsupervised landmark detection and classification of lung infection using transporter neural networks.

Supervised deep learning techniques have been very popular in medical imaging for various tasks of classification, segmentation, and object detection. However, they require a large number of labelled data which is expensive and requires many hours of careful annotation by experts. In this paper, an unsupervised transporter neural network framework with an attention mechanism is proposed to automatically identify relevant landmarks with applications in lung ultrasound (LUS) imaging. The proposed framework identifies key points that provide a concise geometric representation highlighting regions with high structural variation in the LUS videos. In order for the landmarks to be clinically relevant, we have employed acoustic propagation physics driven feature maps and angle-controlled Radon Transformed frames at the input instead of directly employing the gray scale LUS frames. Once the landmarks are identified, the presence of these landmarks can be employed for classification of the given frame into various classes of severity of infection in lung. The proposed framework has been trained on 130 LUS videos and validated on 100 LUS videos acquired from multiple centres at Spain and India. Frames were independently assessed by experts to identify clinically relevant features such as A-lines, B-lines, and pleura in LUS videos. The key points detected showed high sensitivity of 99% in detecting the image landmarks identified by experts. Also, on employing for classification of the given lung image into normal and abnormal classes, the proposed approach, even with no prior training, achieved an average accuracy of 97% and an average F1-score of 95% respectively on the task of co-classification with 3-fold cross-validation.

Unsupervised Multi-Latent Space RL Framework for Video Summarization in Ultrasound Imaging.

The COVID-19 pandemic has highlighted the need for a tool to speed up triage in ultrasound scans and provide clinicians with fast access to relevant information. To this end, we propose a new unsupervised reinforcement learning (RL) framework with novel rewards to facilitate unsupervised learning by avoiding tedious and impractical manual labelling for summarizing ultrasound videos. The proposed framework is capable of delivering video summaries with classification labels and segmentations of key landmarks which enhances its utility as a triage tool in the emergency department (ED) and for use in telemedicine. Using an attention ensemble of encoders, the high dimensional image is projected into a low dimensional latent space in terms of: a) reduced distance with a normal or abnormal class (classifier encoder), b) following a topology of landmarks (segmentation encoder), and c) the distance or topology agnostic latent representation (autoencoders). The summarization network is implemented using a bi-directional long short term memory (Bi-LSTM) which utilizes the latent space representation from the encoder. Validation is performed on lung ultrasound (LUS), that typically represent potential use cases in telemedicine and ED triage acquired from different medical centers across geographies (India and Spain). The proposed approach trained and tested on 126 LUS videos showed high agreement with the ground truth with an average precision of over 80% and average F1 score of well over 44 ±1.7 %. The approach resulted in an average reduction in storage space of 77% which can ease bandwidth and storage requirements in telemedicine.

Neoplastic meningitis: clinico-radiological features, outcome and prognostic factors

Objective: This retrospective hospital based study aimed to describe clinico-radiological features and outcome of neoplastic meningitis (NM) and to evaluate the signifi cance of the presence of malignant cells in CSF and identifi able primary in NM. Methods: The diagnosis of NM was based on the presence of malignant cells in CSF cytology, meningeal biopsy, post mortem examination or compatible clinicoradiological features in patients with known primary malignancy. For subgroup comparisons, Mann Whitney test and Fisher’s exact test were used for continuous and categorical variables respectively. Relative risk of survival in positive CSF cytology for malignant cells and known primary versus negative were calculated. Results: There were 25 patients (mean age 44.5 + 17.6 years) of NM during the study period (2000-2008). They presented with raised ICP headache (72%), cauda equina syndrome (28%), or hemiparesis (12%). Meningeal enhancement and hydrocephalus were seen in 60% and 21% respectively. CSF analysis revealed hypoglychorrachia (64%), raised protein (68%) and pleocytosis (48%). CSF cytology for malignant cells was positive in 76% and cumulative positivity increased by 31% from 1st to 3rd lumbar punctures. A primary could be identifi ed in 56% cases. At last follow up, 16 out of 18 had died. Hypoglychorrachia was the only variable analyzed, which predicted the cytology positivity (p=0.01). The mean duration of survival from the onset was signifi cantly less in cytology positive group (p=0.001). The relative risk of survival at 90 days, 120 and 150 days were signifi cantly higher in cytology and primary negative group compared to positive group. Conclusion: NM with positive cytology or with an identifi able primary tumor has a more aggressive course when compared to the negative groups and former have shorter lifespan. The possibility of positive cytology is high with hypoglychorrachia.