Mission College

Abstract BACKGROUND Progress in automated volumetric segmentation of postoperative glioblastoma has been hindered by limited data availability, primarily related to patient privacy concerns. To overcome this, we developed the Federated Tumor Segmentation (FeTS) 2.0 initiative, which leveraged federated learning to enable training of an AI-based glioblastoma segmentation model by 53 global institutions without the need for data sharing. METHODS Multi-parametric MRI of 10,208 follow up timepoints from postoperative glioblastoma patients (8,100 training, 2,108 hold-out validation) were pre-processed and expertly annotated across the 53 contributing institutions. The provided annotation protocol comprised necrosis (NE), peritumoral edematous/infiltrated tissue (TIE), enhancing tumor (ET), and resection cavity (RC). Federated learning of an nnUNet model, executed through the MLCommons’ MedPerf orchestrator and Intel’s OpenFL framework, proceeded for over 600 epochs with trained model weights and validation scores being the only data shared between collaborators. RESULTS The best performing consensus model achieved hold-out validation Dice Similarity Coefficients (DSC) of 0.95, 0.94, 0.89, and 0.77 for TIE, ET, RC, and NE, respectively, indicating new record performance for postoperative glioblastoma segmentation. In comparison, the centralized BraTS 2024 postoperative glioma study was trained on 1,350 follow up timepoints and tested on 397 hold-out follow up timepoints, with class-specific corresponding DSC of 0.89, 0.79, 0.78, and 0.81 respectively. CONCLUSIONS The FeTS 2.0 initiative demonstrates that federated learning enables scaled, geographically diverse, de-centralized collaboration, expediting big data analysis, while obviating privacy concerns, and hence improves AI performance and generalizability. By enabling reliable and reproducible delineation of postoperative regions-of-interest, our AI model paves the way for large-scale retrospective and prospective analyses aimed at defining volumetric thresholds that correlate with clinical outcomes. Future studies will perform technical and clinical validation on out-of-sample clinical trial datasets, generating evidence to establish three-dimensional cutoff values that can be considered in future RANO criteria, thereby standardizing glioblastoma response assessment and supporting adaptive, data-driven therapeutic decision making worldwide.

Cell-aware diagnostics identify Front-End-Of-Line (FEOL) defects in standard cells, aiding Physical Failure Analysis (PFA) and enhancing root-cause analysis. Our findings show that cell-aware diagnostics often reduce suspect areas to under 1% of the total cell area. Cell faults can be a major failure mechanism given the significant space cells occupy. These benefits led us to create a high-volume manufacturing-friendly scan diagnostics infrastructure using cell-aware diagnostics. We present data-driven solutions for implementing and maintaining this infrastructure. We also include the results of six PFA samples, which validate the FEOL defects across four designs and three technology nodes, yielding six successful results.

Parkinson's disease (PD) is globally the fastestgrowing neurodegenerative disease in the elderly with unclear causes that impairs individuals' motor usage and may induce dementia. Using machine learning technology has the potential to allow for faster and more accurate diagnoses of Parkinson's, which can allow Parkinson's treatment to reach low-income communities with access to fewer health resources. Our study explores a machine learning and deep learning classification algorithm that uses voice recordings to diagnose whether or not a person is likely to have Parkinson's disease. By using various characteristics of the voice recording, such as pitch, frequency, and amplitude, our model is able to accurately predict the presence of Parkinson's with an accuracy of 94%. Unlike other contemporary methods of diagnosing Parkinson's, our research is more accessible as recording voices can be done universally and is not dependent on access to healthcare technology.

Transplanting organs and tissues has become a crucial component of healthcare worldwide. Therefore, it has become imperative to increase public awareness of and attitudes regarding organ donation. Objectives: (i) To ascertain the general understanding of the public regarding organ donation, (ii) To determine how the general public feels about organ donation, (iii) To determine the widespread misconceptions that discourage people from volunteering for organ donation. Methodology: A descriptive cross-sectional questionnaire survey was designed for 100 participants from the general population in Kerala. Results: Overall, 97% of study participants knew something about organ donation and 97% of people expressed support for organ donation. The majority of participants were aware of the laws governing organ donation and its procedure. This study highlights the significance of increasing public knowledge to remove barriers and disbeliefs that keep people from volunteering for organ donation. Knowledge and misconception were found to be interrelated with the general public's attitude towards their willingness to organ donation. Knowledge about the concept of brain death and transplant procedure may help in increasing the rate of willingness towards organ donation.

The study was conducted to assess the effect of planned teaching programme on the knowledge on DASH diet among hypertensive patients of selected wards of JMMC & RI, Thrissur. The objectives of the study were to assess the level of knowledge on DASH diet, to determine the effect of planned teaching programme on DASH diet and to associate socio demographic and clinical data variables with pre-test level knowledge on DASH diet among hypertensive patients in selected wards of JMMC & RI, Thrissur. Quantitative research approach with quasi experimental one group pre-test post-test design was used in the study. The study was conducted in selected wards (cardiology and nephrology) of JMMC &RI, Thrissur. The study population included hypertensive patients above the age of 40 years. Teaching was provided and data collected from 40 samples were analyzed using descriptive and inferential statistics. Major findings of the study were that the overall mean post-test score (26.88) was significantly higher than the mean pre-test knowledge score (16.10). This showed the effectiveness of the planned teaching programme on DASH diet. This finding is supported by a study by Suneesh PM in Unicare hospital at Rajkot. The study was to assess the effectiveness of STP on knowledge regarding DASH diet in controlling blood pressure among hypertensive patients. The mean pre-test knowledge score is 6.63 and the post-test knowledge score is 15. Statistically, there was significant association between the age of the hypertensive patients with the level of knowledge on DASH diet.

A study to assess the effect of structured teaching program on the level of knowledge and practice on administration of insulin among diabetic patients and their care givers in selected wards of JMMC&RI, Thrissur. Objectives: to assess the level of knowledge and practice on administration of insulin among diabetic patients and their care givers before and after intervention, to compare the level of knowledge and practice on administration of insulin among diabetic patients and their caregivers before and after intervention and to associate the pre-test level of knowledge and practice in administration of insulin among diabetes patients and their care givers with selected socio demographic and clinical data variables. Methodology: The study was conducted among 30 diabetes patients on insulin therapy using pre-experimental one group pre-test post-test design. Samples were selected by purposive sampling technique. The tool used for the study had 3 sections: Section A - Socio demographic and clinical data variables where the items include age, gender, marital status, educational status, occupation, monthly income, family history of diabetes mellitus, duration of illness, duration of insulin treatment, person administering insulin, any prior teachings on insulin administration and its source; Section B - Structured knowledge questionnaire; Section C - Structured practice checklist. Result: On data analysis it reveals that 14(46.7%) belongs to the age group of 61-80 years, 19(63.3%) are male, 29(96.7%) are married, 11(36.7%) have high school education, 18(60%) were unemployed‚ 14(46.7%) samples have family history of diabetes mellitus, 13(43.3%) have diabetes for less than 5 years and 4(13.3%) have duration of 11-15 years, 13(43.3%) take insulin for less than 1 year, 14(46.7%) administered insulin by themselves, 23(76.7%) had prior knowledge on insulin administration. The mean post-test knowledge score (19.46) of diabetic patients was higher than the mean pre-test knowledge score (10.2). The computed ‘t’ value (-19.631 ) showed a significant difference between the pre- test and posttest knowledge scores. The mean post-test practice score (19.60) of 30 diabetic patients was found to be significantly higher than their pre-test practice scores. There was a statistically significant relationship between the knowledge and practice level after structured teaching program (P<0.001). There was no significant association between knowledge and practice with selected socio demographic and clinical data variables. The teaching program was accepted by all subjects and showed their interest to learn the administration of insulin.

Silicon photonics is a key element to datacenter connectivity. We review recent progress on fully integrated Silicon photonics transmitters with heterogeneously integrated DFB lasers at Intel Corporation, and discuss future areas of development.

Worldwide COVID-19 pandemic had led to a prolonged stressful condition and bring about more psychological and social effects in the community. This condition is commonly seen in each group and more commonly it has affected in children, older peoples and even the health workers who are exposed and likely to bring about the condition of stress, anxiety, uneasiness and depression. These prolonged psychosocial issues may result in mental health problems and long-term consequence on the mental functioning and coping capacities of family members. The aim of this article is to explain some of the psychosocial impacts, and more is emphasis given in identifying these psychosocial problems and some of the strategies which can be used to overcome these issues.