Developing artificial intelligence systems that are capable of learning at the edge of a network requires both energy-efficient inference and learning. However, current memory technology cannot provide the necessary combination of high endurance, low programming energy and non-destructive read processes. Here we report a unified memory stack that functions as a memristor as well as a ferroelectric capacitor. Memristors are ideal for inference but have limited endurance and high programming energy; ferroelectric capacitors are ideal for learning, but their destructive read process makes them unsuitable for inference. Our memory stack uses a silicon-doped hafnium oxide and titanium scavenging layer that are integrated into the back end of line of a complementary metal–oxide–semiconductor process. With this approach, we fabricate an 18,432-device hybrid array (consisting of 16,384 ferroelectric capacitors and 2,048 memristors) with on-chip complementary metal–oxide–semiconductor periphery circuits. Each weight is associated with an analogue value stored as conductance levels in the memristors and a high-precision hidden value stored as a signed integer in the ferroelectric capacitors. Weight transfers between the different memory technologies occur without a formal digital-to-analogue converter. We use the array to validate an on-chip learning solution that, without batching, performs competitively with floating-point-precision software models across several benchmarks. A memory technology that combines the functions of memristors and ferroelectric capacitors in a single stack can be used for on-chip training and inference of artificial neural networks.
This study evaluates the suitability of shale for ceramic brick production, focusing on mechanical, thermal, and radiological properties. Incorporating 10
Iron deficiency anaemia is the most common cause of anemia among pregnant and lactating women. Personalised management and precision in anemia diagnosis requires ferritin estimation. The current laboratory-based serum ferritin tests are a bottleneck for further management of anemia. The study aimed to assess the diagnostic accuracy of the point-of-care testing (POCT) device for serum ferritin estimation among people in different age groups in a secondary care health facility in India, compared to the gold standard laboratory-based enhanced chemiluminescence-based immunoassay. A sample of 40 participants attending the 50-bed secondary healthcare facility in Faridabad, Haryana, was included. Venous blood samples were collected to estimate serum ferritin levels using the POCT device and the laboratory-based enhanced chemiluminescence-based immunoassay (gold standard). Sensitivity, specificity, positive and negative predictive values, accuracy, and Lin’s concordance correlation coefficient were calculated to assess diagnostic accuracy. This study enrolled 40 participants between the ages of 5 and 45 years, with the majority being females (67.5
We report a case of rhinosinusitis caused by Scedosporium apiospermum in a 35-year-old male who presented with nasal discharge. Diagnosis was made by demonstration of hyaline hyphae in the tissue biopsy and culture characteristics. The patient was managed by surgical debridement and oral voriconazole. The patient responded well to the treatment. This case signifies the importance of culture isolation and antifungal minimal inhibitory concentration (MIC) in each case of fungal rhinosinusitis.
