Using a Deep Learning Model to Explore the Impact of Clinical Data on COVID-19 Diagnosis Using Chest X-ray

The coronavirus pandemic (COVID-19) is disrupting the entire world; its rapid global spread threatens to affect millions of people. Accurate and timely diagnosis of COVID-19 is essential to control the spread and alleviate risk. Due to the promising results achieved by integrating machine learning (ML), particularly deep learning (DL), in automating the multiple disease diagnosis process. In the current study, a model based on deep learning was proposed for the automated diagnosis of COVID-19 using chest X-ray images (CXR) and clinical data of the patient. The aim of this study is to investigate the effects of integrating clinical patient data with the CXR for automated COVID-19 diagnosis. The proposed model used data collected from King Fahad University Hospital, Dammam, KSA, which consists of 270 patient records. The experiments were carried out first with clinical data, second with the CXR, and finally with clinical data and CXR. The fusion technique was used to combine the clinical features and features extracted from images. The study found that integrating clinical data with the CXR improves diagnostic accuracy. Using the clinical data and the CXR, the model achieved an accuracy of 0.970, a recall of 0.986, a precision of 0.978, and an F-score of 0.982. Further validation was performed by comparing the performance of the proposed system with the diagnosis of an expert. Additionally, the results have shown that the proposed system can be used as a tool that can help the doctors in COVID-19 diagnosis.

Synergistic activity of combined inhibition of anti-apoptotic molecules in B-cell precursor ALL

Targeting BCL-2, a key regulator of survival in B-cell malignancies including precursor B-cell acute lymphoblastic leukemia, has become a promising treatment strategy. However, given the redundancy of anti-apoptotic BCL-2 family proteins (BCL-2, BCL-XL, MCL-1), single targeting may not be sufficient. When analyzing the effects of BH3-mimetics selectively targeting BCL-XL and MCL-1 alone or in combination with the BCL-2 inhibitor venetoclax, heterogeneous sensitivity to either of these inhibitors was found in ALL cell lines and in patient-derived xenografts. Interestingly, some venetoclax-resistant leukemias were sensitive to the MCL-1-selective antagonist S63845 and/or BCL-XL-selective A-1331852 suggesting functional mutual substitution. Consequently, co-inhibition of BCL-2 and MCL-1 or BCL-XL resulted in synergistic apoptosis induction. Functional analysis by BH3-profiling and analysis of protein complexes revealed that venetoclax-treated ALL cells are dependent on MCL-1 and BCL-XL, indicating that MCL-1 or BCL-XL provide an Achilles heel in BCL-2-inhibited cells. The effect of combining BCL-2 and MCL-1 inhibition by venetoclax and S63845 was evaluated in vivo and strongly enhanced anti-leukemia activity was found in a pre-clinical patient-derived xenograft model. Our study offers in-depth molecular analysis of mutual substitution of BCL-2 family proteins in acute lymphoblastic leukemia and provides targets for combination treatment in vivo and in ongoing clinical studies.

Increased Expression and Altered Cellular Localization of Fibroblast Growth Factor Receptor-Like 1 (FGFRL1) Are Associated with Prostate Cancer Progression

Fibroblast growth factor receptors (FGFRs) 1–4 are involved in prostate cancer (PCa) regulation, but the role of FGFR-like 1 (FGFRL1) in PCa is unclear. FGFRL1 expression was studied by qRT-PCR and immunohistochemistry of patient tissue microarrays (TMAs) and correlated with clinical patient data. The effects of FGFRL1 knockdown (KD) in PC3M were studied in in vitro culture models and in mouse xenograft tumors. Our results showed that FGFRL1 was significantly upregulated in PCa. The level of membranous FGFRL1 was negatively associated with high Gleason scores (GSs) and Ki67, while increased cytoplasmic and nuclear FGFRL1 showed a positive correlation. Cox regression analysis indicated that nuclear FGFRL1 was an independent prognostic marker for biochemical recurrence after radical prostatectomy. Functional studies indicated that FGFRL1-KD in PC3M cells increases FGFR signaling, whereas FGFRL1 overexpression attenuates it, supporting decoy receptor actions of membrane-localized FGFRL1. In accordance with clinical data, FGFRL1-KD markedly suppressed PC3M xenograft growth. Transcriptomics of FGFRL1-KD cells and xenografts revealed major changes in genes regulating differentiation, ECM turnover, and tumor–stromal interactions associated with decreased growth in FGFRL1-KD xenografts. Our results suggest that FGFRL1 upregulation and altered cellular compartmentalization contribute to PCa progression. The nuclear FGFRL1 could serve as a prognostic marker for PCa patients.

Dynamic PET Imaging Using Dual Texture Features

Purpose: This study aims to explore the impact of adding texture features in dynamic positron emission tomography (PET) reconstruction of imaging results.Methods: We have improved a reconstruction method that combines radiological dual texture features. In this method, multiple short time frames are added to obtain composite frames, and the image reconstructed by composite frames is used as the prior image. We extract texture features from prior images by using the gray level-gradient cooccurrence matrix (GGCM) and gray-level run length matrix (GLRLM). The prior information contains the intensity of the prior image, the inverse difference moment of the GGCM and the long-run low gray-level emphasis of the GLRLM.Results: The computer simulation results show that, compared with the traditional maximum likelihood, the proposed method obtains a higher signal-to-noise ratio (SNR) in the image obtained by dynamic PET reconstruction. Compared with similar methods, the proposed algorithm has a better normalized mean squared error (NMSE) and contrast recovery coefficient (CRC) at the tumor in the reconstructed image. Simulation studies on clinical patient images show that this method is also more accurate for reconstructing high-uptake lesions.Conclusion: By adding texture features to dynamic PET reconstruction, the reconstructed images are more accurate at the tumor.

Nitric Oxide: The Missing Factor in COVID-19 Severity?

Coronavirus disease 2019 (COVID-19) is a contagious respiratory and vascular disease that continues to spread among people around the world, mutating into new strains with increased transmission rates, such as the delta variant. The scientific community is struggling to discover the link between negative COVID-19 outcomes in patients with preexisting conditions, as well as identify the cause of the negative clinical patient outcomes (patients who need medical attention, including hospitalization) in what seems like a widespread range of COVID-19 symptoms that manifest atypically to any preexisting respiratory tract infectious diseases known so far. Having successfully developed a nutritional formulation intervention based on nitrate, a nitric oxide precursor, the authors hypothesis is that both the comorbidities associated with negative clinical patient outcomes and symptoms associated with COVID-19 sickness are linked to the depletion of a simple molecule: nitric oxide.

Professional decision making with digitalisation of patient contacts in a medical advice setting: a qualitative study of a pilot project with a chat programme in Sweden

ObjectivesPatient e-services are increasingly launched globally to make healthcare more efficient and digitalised. One area that is digitalised is medical advice, where patients asynchronously chat with nurses and physicians, with patients having filled in a form with predefined questions before the chat. This study aimed to explore how occupational professionalism and the possibility of professional judgement are affected when clinical patient contact is digitalised. The study’s overall question concerns whether and how the scope of the healthcare staff’s professional judgement and occupational professionalism are affected by digitalisation.Design and settingA qualitative study of healthcare professionals working in a pilot project with a chat programme for patients in a medical advice setting in Sweden.Participants and analysisContextual inquiries and 17 interviews with nurses (n=9) and physicians (n=8). The interviews were thematically analysed. The analysis was inductive and based on theories of decision making.ResultsThree themes emerged: (1) Predefined questions to patients not tailored for healthcare professionals’ work, (2) reduced trust in written communication and (3) reduced opportunity to obtain information through chat communication.ConclusionsThe results indicate that asynchronous chat with patients might reduce the opportunity for nurses and physicians to obtain and use professional knowledge and discretionary decision making. Furthermore, the system’s design increases uncertainty in assessments and decision making, which reduces the range of occupational professionalism.

OP123 The Use Of Surrogate Outcomes In National Institute For Health And Care Excellence (NICE) Highly Specialised Technology Evaluations: A Review Of Published Guidance

IntroductionThe use of surrogate outcomes in health technology assessment (HTA) is increasing and methods for validating surrogate relationships have been published. However, these may not be fully applicable to ultra-rare diseases due to challenges such as scarcity of evidence and heterogenous populations. This study reviews and summarizes the use of surrogate outcomes and committee's considerations in the evaluations within the National Institute for Health and Care Excellence's (NICE) Highly Specialised Technology (HST) programme, which was established in 2013 in response to the challenges associated with the assessment of ultra-rare diseases.MethodsAll HST evaluation documents published before November 2020 were reviewed. Data extracted included surrogate outcomes used, rationales, the committee's considerations on the validity and generalizability of the surrogate relationships, related uncertainties, and other factors considered in decision-making.ResultsSeven out of the eighteen published HST topics used surrogate outcomes. The rationale for most of the surrogate relationships focused on biological plausibility. Common concerns raised by the committee included the generalizability of the surrogate relationship to the condition of interest, the lack of validation, and inability to prove or quantify the magnitude of benefits associated with the surrogate relationships. In some topics, other aspects of the evidence and clinical/patient expert's opinions were also considered by the committee.ConclusionsThe use of surrogate outcomes is common in NICE HST evaluations and the challenges in supporting surrogate relationships with more than biological plausibility are recognized. However, our review indicates that, the committee considers more than just biological plausibility and will take into account other related factors.

Meningitis, meningoencephalitis and encephalitis in Bern: an observational study of 258 patients

Background Depending on geographic location, causes of encephalitis, meningoencephalitis and meningitis vary substantially. We aimed to identify the most frequent causes, clinical presentation and long-term outcome of encephalitis, meningoencephalitis and meningitis cases treated in the Inselspital University Hospital Bern, Switzerland. Methods In this monocentric, observational study, we performed a retrospective review of clinical patient records for all patients treated within a 3-year period. Patients were contacted for a telephone follow-up interview and to fill out questionnaires, especially related to disturbances of sleep and wakefulness. Results We included 258 patients with the following conditions: encephalitis (18%), nonbacterial meningoencephalitis (42%), nonbacterial meningitis (27%) and bacterial meningoencephalitis/meningitis (13%). Herpes simplex virus (HSV) was the most common cause of encephalitis (18%); tick-borne encephalitis virus (TBEV) was the most common cause of nonbacterial meningoencephalitis (46%), enterovirus was the most common cause of nonbacterial meningitis (21%) and Streptococcus pneumoniae was the most common cause of bacterial meningoencephalitis/meningitis (49%). Overall, 35% patients remained without a known cause. After a median time of 16 months, 162 patients participated in the follow-up interview; 56% reported suffering from neurological long-term sequelae such as fatigue and/or excessive daytime sleepiness (34%), cognitive impairment and memory deficits (22%), headache (14%) and epileptic seizures (11%). Conclusions In the Bern region, Switzerland, TBEV was the overall most frequently detected infectious cause, with a clinical manifestation of meningoencephalitis in the majority of cases. Long-term neurological sequelae, most importantly cognitive impairment, fatigue and headache, were frequently self-reported not only in encephalitis and meningoencephalitis survivors but also in viral meningitis survivors up to 40 months after acute infection.

Localized alopecia and suppression of hypothalamic-pituitary-adrenal (HPA) axis in dogs following treatment with difluprednate 0.05% ophthalmic emulsion (Durezol®)

Background Despite the common use of topical ophthalmic corticosteroids in dogs, detailed reports on systemic and dermatologic adverse effects are limited. Results Nine purpose-bred research Beagles were treated with difluprednate 0.05% ophthalmic emulsion in one or both eyes 2–3 times daily. Some difluprednate treated dogs developed mild to severe alopecia of the periocular region, face, and distal pinna (5/9). The median duration of treatment prior to onset of dermatologic signs for difluprednate treated dogs was 550 days (453–1160 days). Diagnostic testing included complete blood count (CBC) and serum biochemistry, adrenocorticotropic hormone (ACTH) stimulation testing combined with endogenous ACTH measurement, and skin biopsy. The CBC and chemistry were within normal limits for all dogs. There were varying degrees of suppression of the hypothalamic-pituitary-adrenocortical (HPA) axis with difluprednate treatment. Dogs with the most profound alopecic changes had less pronounced HPA axis suppression compared to dogs with no integumentary changes. Skin biopsies demonstrated follicular atrophy and follicular keratosis. When topical difluprednate was reduced to unilateral therapy, the hair regrew on the untreated side of the face. In addition to the affected research dogs, a 7-year old female spayed Chihuahua that was being treated as a clinical patient with long-term difluprednate 0.05% ophthalmic emulsion developed generalized hypotrichosis on the head and body and a potbellied appearance. ACTH stimulation testing revealed suppression of the HPA axis with a mild increase in serum alkaline phosphatase (ALP) activity and a urine specific gravity of 1.016. The combination of clinical signs and laboratory abnormalities was supportive of iatrogenic hyperadrenocorticism. Conclusions In dogs long-term use of difluprednate ophthalmic emulsion results in HPA axis suppression and in some cases iatrogenic hyperadrenocorticism. A novel pattern of localized alopecia is suspected to be related to dermal absorption and local action due to superior potency and penetration compared to other commonly utilized ophthalmic corticosteroids.