An ASER AI/ML expert panel formative user research study for an interpretable interactive splenic AAST grading graphical user interface prototype.

PURPOSE: The AAST Organ Injury Scale is widely adopted for splenic injury severity but suffers from only moderate inter-rater agreement. This work assesses SpleenPro, a prototype interactive explainable artificial intelligence/machine learning (AI/ML) diagnostic aid to support AAST grading, for effects on radiologist dwell time, agreement, clinical utility, and user acceptance. METHODS: Two trauma radiology ad hoc expert panelists independently performed timed AAST grading on 76 admission CT studies with blunt splenic injury, first without AI/ML assistance, and after a 2-month washout period and randomization, with AI/ML assistance. To evaluate user acceptance, three versions of the SpleenPro user interface with increasing explainability were presented to four independent expert panelists with four example cases each. A structured interview consisting of Likert scales and free responses was conducted, with specific questions regarding dimensions of diagnostic utility (DU); mental support (MS); effort, workload, and frustration (EWF); trust and reliability (TR); and likelihood of future use (LFU). RESULTS: SpleenPro significantly decreased interpretation times for both raters. Weighted Cohen's kappa increased from 0.53 to 0.70 with AI/ML assistance. During user acceptance interviews, increasing explainability was associated with improvement in Likert scores for MS, EWF, TR, and LFU. Expert panelists indicated the need for a combined early notification and grading functionality, PACS integration, and report autopopulation to improve DU. CONCLUSIONS: SpleenPro was useful for improving objectivity of AAST grading and increasing mental support. Formative user research identified generalizable concepts including the need for a combined detection and grading pipeline and integration with the clinical workflow.

A survey of ASER members on artificial intelligence in emergency radiology: trends, perceptions, and expectations.

PURPOSE: There is a growing body of diagnostic performance studies for emergency radiology-related artificial intelligence/machine learning (AI/ML) tools; however, little is known about user preferences, concerns, experiences, expectations, and the degree of penetration of AI tools in emergency radiology. Our aim is to conduct a survey of the current trends, perceptions, and expectations regarding AI among American Society of Emergency Radiology (ASER) members. METHODS: An anonymous and voluntary online survey questionnaire was e-mailed to all ASER members, followed by two reminder e-mails. A descriptive analysis of the data was conducted, and results summarized. RESULTS: A total of 113 members responded (response rate 12%). The majority were attending radiologists (90%) with greater than 10 years' experience (80%) and from an academic practice (65%). Most (55%) reported use of commercial AI CAD tools in their practice. Workflow prioritization based on pathology detection, injury or disease severity grading and classification, quantitative visualization, and auto-population of structured reports were identified as high-value tasks. Respondents overwhelmingly indicated a need for explainable and verifiable tools (87%) and the need for transparency in the development process (80%). Most respondents did not feel that AI would reduce the need for emergency radiologists in the next two decades (72%) or diminish interest in fellowship programs (58%). Negative perceptions pertained to potential for automation bias (23%), over-diagnosis (16%), poor generalizability (15%), negative impact on training (11%), and impediments to workflow (10%). CONCLUSION: ASER member respondents are in general optimistic about the impact of AI in the practice of emergency radiology and its impact on the popularity of emergency radiology as a subspecialty. The majority expect to see transparent and explainable AI models with the radiologist as the decision-maker.

Using a Model-based Material Sparing Approach for Formulation and Process Development of a Roller Compacted Drug Product.

The present work details a material sparing approach that combines material profiling with Instron uniaxial die-punch tester and use of a roller compaction mathematical model to guide both formulation and process development of a roller-compacted drug product. True density, compression profiling, and frictional properties of the pre-blend powders are used as inputs for the predictive roller compaction model, while flow properties, particle size distribution, and assay uniformity of roller compaction granules are used to select formulation composition and ribbon solid fraction. Using less than 10 g of a model drug compound for material profiling, roller compacted blend in capsule formulations with appropriate excipient ratios were developed at both 1.4% and 14.4% drug loadings. Subsequently, scale-up batches were successfully manufactured based on the roller compaction process parameters obtained from predictive modeling. The measured solid fractions of roller compaction ribbon samples from the scale-up batches were in good agreement with the target solid fraction of the modeling. This approach demonstrated considerable advantages through savings in both materials and number of batches in the development of a roller-compacted drug product, which is of particular value at early development stages when drug substance is often limited and timelines are aggressive.

Multi-task driven explainable diagnosis of COVID-19 using chest X-ray images.

With increasing number of COVID-19 cases globally, all the countries are ramping up the testing numbers. While the RT-PCR kits are available in sufficient quantity in several countries, others are facing challenges with limited availability of testing kits and processing centers in remote areas. This has motivated researchers to find alternate methods of testing which are reliable, easily accessible and faster. Chest X-Ray is one of the modalities that is gaining acceptance as a screening modality. Towards this direction, the paper has two primary contributions. Firstly, we present the COVID-19 Multi-Task Network (COMiT-Net) which is an automated end-to-end network for COVID-19 screening. The proposed network not only predicts whether the CXR has COVID-19 features present or not, it also performs semantic segmentation of the regions of interest to make the model explainable. Secondly, with the help of medical professionals, we manually annotate the lung regions and semantic segmentation of COVID19 symptoms in CXRs taken from the ChestXray-14, CheXpert, and a consolidated COVID-19 dataset. These annotations will be released to the research community. Experiments performed with more than 2500 frontal CXR images show that at 90% specificity, the proposed COMiT-Net yields 96.80% sensitivity.

Utility of contemporaneous dual read in the setting of emergency teleradiology reporting.

PURPOSE: Emergency radiology requires rapid and accurate interpretation of imaging examinations. Missed findings may lead to adverse outcomes. Double reporting may be used to minimize errors. Limited contemporaneous double reporting may be most efficient and cost-effective, but no data exists. This study is intended to examine the benefits of double reading and identify examinations where this would be most useful. METHODS: In this study, dual reporting was conducted in a parallel reading environment in a teleradiology practice for 3779 radiological procedures performed at two radiology centers in the USA over a period of 4 months. Discrepancies between reads were scored using the ACR peer review scoring system and grouped by modality and body part. Errors were tabulated across the study types, followed by identification of statistically significant differences. The interaction between image number and odds of an error was ascertained. RESULTS: In 145 instances (3.8%; 95 % CI, 3.2-4.4%), double reporting identified errors, leading to report modification. Study type was significantly related to error frequency (p = 0.0001), with higher than average frequencies of error seen for CT abdomen and pelvis and MRI head or spine, but lower than average for CT head, CT spine, and ultrasound. Image number was positively associated with error odds, but was not independently significant in a joint logistic regression model that included study type. CONCLUSION: Dual reporting identifies missed findings in about 1 of 25 emergency studies. This benefit varies substantially across study types and limited double reporting, merits further investigation as a cost-effective practice improvement strategy.

Nucleating emergency radiology specialization in India.

Emergency radiology is being recognized as an important and distinct specialty of radiology which merits utmost attention of educators, radiology program curricula committees, and radiology practices in India. Providing an accurate but timely diagnosis requires a skilled judgement and a strong process framework, particularly in acute trauma setting or a life-threatening acute illness. However, due to a shortage of radiologists in India and lack of awareness and suitable opportunities, there has been no concerted movement towards emergency radiology subspecialty training or dedicated emergency radiology positions. It was with these gaps in mind that the Society for Emergency Radiology was envisioned in 2012 and formulated in 2013. The proposed role of the Society for Emergency Radiology is to identify deficiencies in the field, namely, lack of adequate exposure, lack of mentorship by experienced emergency radiologists, lack of suitable opportunities for emergency radiologists; establish standards of practice; and promote education and implementation research to bridge the gaps. Through collaboration with other societies and partnership with the journal Emergency Radiology, the Society for Emergency Radiology hopes to promote a free exchange of ideas, protocols, and multi-institutional trials across continents.

Hot Melt Extrusion: Development of an Amorphous Solid Dispersion for an Insoluble Drug from Mini-scale to Clinical Scale.

The objective of the study was to develop an amorphous solid dispersion (ASD) for an insoluble compound X by hot melt extrusion (HME) process. The focus was to identify material-sparing approaches to develop bioavailable and stable ASD including scale up of HME process using minimal drug. Mixtures of compound X and polymers with and without surfactants or pH modifiers were evaluated by hot stage microscopy (HSM), polarized light microscopy (PLM), and modulated differential scanning calorimetry (mDSC), which enabled systematic selection of ASD components. Formulation blends of compound X with PVP K12 and PVP VA64 polymers were extruded through a 9-mm twin screw mini-extruder. Physical characterization of extrudates by PLM, XRPD, and mDSC indicated formation of single-phase ASD's. Accelerated stability testing was performed that allowed rapid selection of stable ASD's and suitable packaging configurations. Dissolution testing by a discriminating two-step non-sink dissolution method showed 70-80% drug release from prototype ASD's, which was around twofold higher compared to crystalline tablet formulations. The in vivo pharmacokinetic study in dogs showed that bioavailability from ASD of compound X with PVP VA64 was four times higher compared to crystalline tablet formulations. The HME process was scaled up from lab scale to clinical scale using volumetric scale up approach and scale-independent-specific energy parameter. The present study demonstrated systematic development of ASD dosage form and scale up of HME process to clinical scale using minimal drug (∼500 g), which allowed successful clinical batch manufacture of enabled formulation within 7 months.

Development of Tablet Formulation of Amorphous Solid Dispersions Prepared by Hot Melt Extrusion Using Quality by Design Approach.

The objective of the study was to identify the extragranular component requirements (level and type of excipients) to develop an immediate release tablet of solid dispersions prepared by hot melt extrusion (HME) process using commonly used HME polymers. Solid dispersions of compound X were prepared using polyvinyl pyrrolidone co-vinyl acetate 64 (PVP VA64), Soluplus, and hypromellose acetate succinate (HPMCAS-LF) polymers in 1:2 ratio by HME through 18 mm extruder. A mixture design was employed to study effect of type of polymer, filler (microcrystalline cellulose (MCC), lactose, and dicalcium phosphate anhydrous (DCPA)), and disintegrant (Crospovidone, croscarmellose sodium, and sodium starch glycolate (SSG)) as well as level of extrudates, filler, and disintegrant on tablet properties such as disintegration time (DT), tensile strength (TS), compactibility, and dissolution. Higher extrudate level resulted in longer DT and lower TS so 60-70% was the maximum amount of acceptable extrudate level in tablets. Fast disintegration was achieved with HPMCAS-containing tablets, whereas Soluplus- and PVP VA64-containing tablets had higher TS. Crospovidone and croscarmellose sodium were more suitable disintegrant than SSG to achieve short DT, and MCC was a suitable filler to prepare tablets with acceptable TS for each studied HME polymer. The influence of extragranular components on dissolution from tablets should be carefully evaluated while finalizing tablet composition, as it varies for each HME polymer. The developed statistical models identified suitable level of fillers and disintegrants for each studied HME polymer to achieve tablets with rapid DT (<15 min) and acceptable TS (≥1 MPa at 10-15% tablet porosity), and their predictivity was confirmed by conducting internal and external validation studies.

Digital transformation of career landscapes in radiology: personal and professional implications.

Millennial radiology is marked by technical disruptions. Advances in internet, digital communications and computing technology, paved way for digitalized workflow orchestration of busy radiology departments. The COVID pandemic brought teleradiology to the forefront, highlighting its importance in maintaining continuity of radiological services, making it an integral component of the radiology practice. Increasing computing power and integrated multimodal data are driving incorporation of artificial intelligence at various stages of the radiology image and reporting cycle. These have and will continue to transform the career landscape in radiology, with more options for radiologists with varied interests and career goals. The ability to work from anywhere and anytime needs to be balanced with other aspects of life. Robust communication, internal and external collaboration, self-discipline, and self-motivation are key to achieving the desired balance while practicing radiology the unconventional way.

Type 2 Diabetes Mellitus in Patients With Polycystic Ovary Syndrome.

Polycystic ovary syndrome (PCOS) is a multisystemic disorder usually seen in females who are in their reproductive age (15-49 years of age). PCOS exhibits insulin resistance and hyperinsulinemia, which make it a pre-diabetic state. The syndrome has many overt changes, like dyslipidemia and hypertension, which increase the risk of cardiovascular diseases. There is also an increased risk of development of hepatic steatosis. Resistance to insulin, increased amount of insulin, and dysfunction of beta-cells are frequent in PCOS, although they are not the only cause for diagnosis. Type 2 diabetes and glucose resistance may result from total or compared insulin insufficiency, which can happen if the beta cells' compensatory response slows down. Pregnancy challenges such as miscarriage, gestational diabetes mellitus (DM), hypertensive disorders of pregnancy, more excellent rates of cesarean birth, and abnormalities in fetal development may be more common in women with PCOS. In studies investigating the glucose-insulin system compared to control groups with similar age and weight, glycemic intolerance, which includes both decreased glucose tolerance and type 2 diabetes, was more common in PCOS women. In the short-term therapy of insulin resistance in PCOS, the potential use of insulin-sensitizing medications has recently been studied. Controlled studies have demonstrated that metformin treatment can lower fasting and stimulate plasma insulin levels by encouraging body weight reduction. These findings provide insulin-sensitizing drugs as a unique method in treating ovarian hyperandrogenism and irregular ovulation in PCOS and indicate a new prescription for Metformin. They further assert that long-term metformin treatment may assist in addressing insulin resistance, reducing the risk of type 2 diabetes and cardiovascular-related disease in people who take it.

Emergency Teleradiology-Past, Present, and, Is There a Future?

Emergency radiology has evolved into a distinct radiology subspecialty requiring a specialized skillset to make a timely and accurate diagnosis of acutely and critically ill or traumatized patients. The need for emergency and odd hour radiology coverage fuelled the growth of internal and external teleradiology and the "nighthawk" services to meet the increasing demands from all stakeholders and support the changing trends in emergency medicine and trauma surgery inclined toward increased reliance on imaging. However, the basic issues of increased imaging workload, radiologist demand-supply mismatch, complex imaging protocols are only partially addressed by teleradiology with the promise of workload balancing by operations to scale. Incorporation of artificially intelligent tools helps scale manifold by the promise of streamlining the workflow, improved detection and quantification as well as prediction. The future of emergency teleradiologists and teleradiology groups is entwined with their ability to incorporate such tools at scale and adapt to newer workflows and different roles. This agility to adopt and adapt would determine their future.

Semantic segmentation of bone structures in chest X-rays including unhealthy radiographs: A robust and accurate approach.

The chest X-ray is a widely used medical imaging technique for the diagnosis of several lung diseases. Some nodules or other pathologies present in the lungs are difficult to visualize on chest X-rays because they are obscured byoverlying boneshadows. Segmentation of bone structures and suppressing them assist medical professionals in reliable diagnosis and organ morphometry. But segmentation of bone structures is challenging due to fuzzy boundaries of organs and inconsistent shape and size of organs due to health issues, age, and gender. The existing bone segmentation methods do not report their performance on abnormal chest X-rays, where it is even more critical to segment the bones. This work presents a robust encoder-decoder network for semantic segmentation of bone structures on normal as well as abnormal chest X-rays. The novelty here lies in combining techniques from two existing networks (Deeplabv3+ and U-net) to achieve robust and superior performance. The fully connected layers of the pre-trained ResNet50 network have been replaced by an Atrous spatial pyramid pooling block for improving the quality of the embedding in the encoder module. The decoder module includes four times upsampling blocks to connect both low-level and high-level features information enabling us to retain both the edges and detail information of the objects. At each level, the up-sampled decoder features are concatenated with the encoder features at a similar level and further fine-tuned to refine the segmentation output. We construct a diverse chest X-ray dataset with ground truth binary masks of anterior ribs, posterior ribs, and clavicle bone for experimentation. The dataset includes 100 samples of chest X-rays belonging to healthy and confirmed patients of lung diseases to maintain the diversity and test the robustness of our method. We test our method using multiple standard metrics and experimental results indicate an excellent performance on both normal and abnormal chest X-rays.

Superiority of Mesoporous Silica-Based Amorphous Formulations over Spray-Dried Solid Dispersions.

The aim of this study was to compare the performance of two amorphous formulation strategies: mesoporous silica via solvent impregnation, and solid dispersions by spray drying. Poorly soluble fenofibrate was chosen as the model drug compound. A total of 30% Fenofibrate-loaded mesoporous silica and spray-dried solid dispersions (SDD) were prepared for head-to-head comparisons, including accelerated stability, manufacturability, and in vitro biorelevant dissolution. In the accelerated stability study under 40 °C/75% RH in open dish, mesoporous silica was able to maintain amorphous fenofibrate for up to 3 months based on solid-state characterizations by PXRD and DSC. This result was superior compared to SDD, as recrystallization was observed within 2 weeks. Under the same drug load, fenofibrate-loaded mesoporous silica showed much better flowability than fenofibrate-loaded SDD, which is beneficial for powder handling of the intermediate product during the downstream process. The in vitro 2-stage dissolution results indicated a well-controlled release of fenofibrate from mesoporous silica in the biorelevant media, rather than a burst release followed by fast precipitation due to the recrystallization in the early simulated gastric phase for SDD. The present study demonstrates that mesoporous silica is a promising formulation platform alternative to prevailing spray-dried solid dispersions for oral drug product development.

The Current States, Challenges, Ongoing Efforts, and Future Perspectives of Pharmaceutical Excipients in Pediatric Patients in Each Country and Region.

A major hurdle in pediatric formulation development is the lack of safety and toxicity data on some of the commonly used excipients. While the maximum oral safe dose for several kinds of excipients is known in the adult population, the doses in pediatric patients, including preterm neonates, are not established yet due to the lack of evidence-based data. This paper consists of four parts: (1) country-specific perspectives in different parts of the world (current state, challenges in excipients, and ongoing efforts) for ensuring the use of safe excipients, (2) comparing and contrasting the country-specific perspectives, (3) past and ongoing collaborative efforts, and (4) future perspectives on excipients for pediatric formulation. The regulatory process for pharmaceutical excipients has been developed. However, there are gaps between each region where a lack of information and an insufficient regulation process was found. Ongoing efforts include raising issues on excipient exposure, building a region-specific database, and improving excipient regulation; however, there is a lack of evidence-based information on safety for the pediatric population. More progress on clear safety limits, quantitative information on excipients of concern in the pediatric population, and international harmonization of excipients' regulatory processes for the pediatric population are required.

Systematic survey of discrepancy rates in an international teleradiology service.

International teleradiology services (ITS) to the United States are based on the principle of deploying American board-certified radiologists across global time zones to optimally distribute the workload. While errors may be reduced by circumventing the traditional night call, there is limited evidence on the actual error rates of teleradiology groups. We have a comprehensive quality assurance (QA) process in our practice, which includes a review of discrepancies between preliminary reports and the final reports by the on-site radiologists. We analyzed the discrepancy QA data to determine the error rates. Archived QA data for 126,449 cases over a period of 1 year (2008) were analyzed for the discrepancy rate, nature of errors, and possible contributory factors. The scores ranged from 0 (no error) to 5 (clinically significant in the acute setting) based on the level of clinical significance. A novel modified Lorenz plot was used to estimate the degree of underreporting and to estimate the true error rate. An internal review of 200 cases was performed to validate the findings. Of the total, there was a total of 227 confirmed errors (0.18%, 95% CI, 0.16 to 0.20). Of these, the majority were levels 2 and 3 (minor error and error of long-term significance but not in the acute setting). Even after correction for underreporting, error rates were less than 1% for clinically significant errors. ITS is associated with very low rates of clinically significant errors. Due to limited feedback, particularly for minor errors, an internal review is important.

Trends in COVID-19 Publications: Streamlining Research Using NLP and LDA.

Background: Research publications related to the novel coronavirus disease COVID-19 are rapidly increasing. However, current online literature hubs, even with artificial intelligence, are limited in identifying the complexity of COVID-19 research topics. We developed a comprehensive Latent Dirichlet Allocation (LDA) model with 25 topics using natural language processing (NLP) techniques on PubMed® research articles about "COVID." We propose a novel methodology to develop and visualise temporal trends, and improve existing online literature hubs. Our results for temporal evolution demonstrate interesting trends, for example, the prominence of "Mental Health" and "Socioeconomic Impact" increased, "Genome Sequence" decreased, and "Epidemiology" remained relatively constant. Applying our methodology to LitCovid, a literature hub from the National Center for Biotechnology Information, we improved the breadth and depth of research topics by subdividing their pre-existing categories. Our topic model demonstrates that research on "masks" and "Personal Protective Equipment (PPE)" is skewed toward clinical applications with a lack of population-based epidemiological research.

Best practices for selection of excipients for paediatrics - Workshop reflection.

The development of age appropriate formulations for the paediatric population has become one of the key areas of focus for the pharmaceutical industry - with a subsequent influence on excipient use. Selection of excipients with appropriate safety and tolerability is a major hurdle in paediatric formulation development. Various factors influence selection of excipients, including target age group, route of administration and dosage form. Evaluation of these factors and a clear rationale and justification is expected by the regulators when it comes to selecting excipients for paediatric formulation. Scientists are encouraged to apply the principle of benefit to risk balance to assess the suitability of excipients to the specific paediatric population for whom the formulation is intended. In order to understand how scientists approach the task of establishing the risk to benefit analysis, a workshop was organised by the European Paediatric Formulation Initiative (EuPFI) to reflect on the current scenario and the different practices employed by formulation scientists in the selection of excipients for paediatric formulations. Aspects assessed by regulators were also canvassed. Finally, the participants were asked to comment on how selecting excipients for use in paediatric formulations may differ from the considerations applied in selecting excipients for formulations for other age groups. Based on the workshop discussion, some recommendations and questions to consider emerged regarding the selection of excipients in paediatric drug development. These best practice recommendations provided a good starting point for a more systematic strategy for selecting excipients for paediatric formulation development.

A Novel Use of Nanocrystalline Suspensions to Develop Sub-Microgram Dose Micro-Tablets.

Developing solid oral drug products with good content uniformity (CU) at low doses is challenging; this challenge further aggravates when the tablet size decreases from a conventional tablet to a micro/mini-tablet (1.2-3 mm diameter). To alleviate the CU issues, we present a novel use of nanocrystalline suspension combined with high shear wet granulation for the first time. In this approach, nanomilled drug in the form of nanocrystalline suspension is sprayed onto the powder bed to ensure uniform distribution. The resulting granules had adequate particle size distribution and flow characteristics to enable manufacturing of micro-tablets with good weight uniformity and tensile strength. Nanomilled drug resulted in excellent content uniformity among individual micro-tablets even at a dose strength as low as 0.16 mcg, whereas micronized drug resulted in unacceptable CU even at 5x higher dose strength (0.8 mcg). Besides, the use of nanomilled drug has enhanced the dosing flexibility of micro-tablets and showed superior dissolution performance in comparison with micronized drug with no impact of storage conditions (40 °C/75%RH for six months) on their dissolution performance. The proposed approach is simple and can be easily incorporated into traditional high shear wet granulation process to develop sub-microgram dose solid oral drug products.

Pharmaceutical amorphous solid dispersion: A review of manufacturing strategies.

Amorphous solid dispersions (ASDs) are popular for enhancing the solubility and bioavailability of poorly water-soluble drugs. Various approaches have been employed to produce ASDs and novel techniques are emerging. This review provides an updated overview of manufacturing techniques for preparing ASDs. As physical stability is a critical quality attribute for ASD, the impact of formulation, equipment, and process variables, together with the downstream processing on physical stability of ASDs have been discussed. Selection strategies are proposed to identify suitable manufacturing methods, which may aid in the development of ASDs with satisfactory physical stability.

Validation of expert system enhanced deep learning algorithm for automated screening for COVID-Pneumonia on chest X-rays.

SARS-CoV2 pandemic exposed the limitations of artificial intelligence based medical imaging systems. Earlier in the pandemic, the absence of sufficient training data prevented effective deep learning (DL) solutions for the diagnosis of COVID-19 based on X-Ray data. Here, addressing the lacunae in existing literature and algorithms with the paucity of initial training data; we describe CovBaseAI, an explainable tool using an ensemble of three DL models and an expert decision system (EDS) for COVID-Pneumonia diagnosis, trained entirely on pre-COVID-19 datasets. The performance and explainability of CovBaseAI was primarily validated on two independent datasets. Firstly, 1401 randomly selected CxR from an Indian quarantine center to assess effectiveness in excluding radiological COVID-Pneumonia requiring higher care. Second, curated dataset; 434 RT-PCR positive cases and 471 non-COVID/Normal historical scans, to assess performance in advanced medical settings. CovBaseAI had an accuracy of 87% with a negative predictive value of 98% in the quarantine-center data. However, sensitivity was 0.66-0.90 taking RT-PCR/radiologist opinion as ground truth. This work provides new insights on the usage of EDS with DL methods and the ability of algorithms to confidently predict COVID-Pneumonia while reinforcing the established learning; that benchmarking based on RT-PCR may not serve as reliable ground truth in radiological diagnosis. Such tools can pave the path for multi-modal high throughput detection of COVID-Pneumonia in screening and referral.