Clinical Acceptability of Artificial Intelligence-Screened Interstitial Lung Disease (AI-ILD) in Lung Cancer Patients Treated with Radiotherapy

PURPOSE/OBJECTIVE(S)

Patients with interstitial lung disease (ILD) treated with thoracic radiotherapy (RT) are at greater risk of pulmonary toxicity. Automatic universal screening for ILD allows radiation oncologists (ROs) to risk stratify patients and implement necessary modifications to their respiratory monitoring or treatment. Automatic screening however may affect RO workload and so it is imperative to assess the clinical acceptability of this tool.

MATERIALS/METHODS

We have developed a machine learning algorithm to identify patients who are at high risk of having ILD based on RT planning computed tomography (CT) images. A quality improvement (QI) project was initiated to test feasibility and acceptability of the machine learning algorithm. If positive, the results of the machine learning algorithm were made available to ROs via structured electronic reporting. ROs were prompted to review the patient and consider expert radiologist consultation if thought appropriate. All electronic surveys and qualitative comments were summarized to describe clinical acceptability. Expert radiologist established gold standard ILD status of all patients on the study. A formal review of RO feedback was collected for all screen-positive, true-positive cases.

RESULTS

Two hundred forty cases were screened of which 45 were flagged as AI-ILD positive and the responsible RO notified. Of these 45 screen-positive cases, all continued on to RT except for 3 patients with tumor progression. From these 45, 24 surveys were completed, 21 had no prior suspicion of ILD. There were 7 true-positives, of which 1 had a survey response. Based on the survey responses, 88% of cases underwent review by the responsible RO. In 16 cases this automatic notification prompted case consultation with an expert radiologist. Expert review was performed from 10 minutes up to 53 hours after the email prompt to the radiologist, with median response time of 1.5 hours. In the 7 screen-positive, true-positive cases, only 2 were not previously known to the responsible RO. In the two cases where true-positive ILD status was previously unknown, one was a mild case of ILD and the other had previously received thoracic RT at this institution without ILD being identified, in both cases the ROs were grateful that this diagnosis was identified prior to treatment. RO confidence in the machine learning prediction was moderate due to the high proportion of false positives.

CONCLUSION

Based on available survey results, more than 75% of the screen-positive cases were reviewed by the responsible RO and two-thirds of these involved expert radiology input. RO feedback was generally positive and this tool was rated as a net benefit despite the high rate of false-positives and the need for clarification.

Association of Artificial Intelligence-Screened Interstitial Lung Disease with Radiation Pneumonitis and Mortality in Locally Advanced Non-Small Cell Lung Cancer

PURPOSE/OBJECTIVE(S)

Radiation pneumonitis (RP) is a common and dose-limiting toxicity following radiotherapy for non-small cell lung cancer (NSCLC). Patients with interstitial lung disease (ILD) are believed to be at increased risk of developing complications including RP, ILD progression, or death. An automated method to identify patients prior to radiotherapy at high risk of developing toxicities or death may allow clinicians to mitigate risk through informed treatment planning and careful patient monitoring.

MATERIALS/METHODS

All locally advanced NSCLC patients treated with definitive radiation from 2006-2021 with a minimum 1 year of follow-up were assessed. RP and mortality data were prospectively collected and retrospectively reviewed. A convolutional neural network (CNN) was previously developed and validated to identify patients with radiographic ILD using planning computed tomography (CT) images, with an accuracy of 0.82. Planning CT scans for the retrospective cohort were used as input to the CNN, with artificial intelligence-screened ILD (AI-ILD) score as an output. AI-ILD scores above our established threshold were labeled as AI-ILD+. The association between AI-ILD score, AI-ILD+/-, mean lung dose (MLD), and the primary outcome of grade ≥2 (G2+) RP or mortality, as well as the secondary outcomes of G2+ RP and mortality were assessed using Wilcoxon rank sum test, univariate and multivariable logistic regression, and Kaplan-Meier survival analysis.

RESULTS

Of 799 patients reviewed, 745 eligible patients were included in the analysis; grade 0-5 RP was reported in 51.3%, 27.1%, 16.9%, 4.0%, 0.1%, and 0.5% of patients respectively. Overall, 22.9% of patients were AI-ILD+, and therefore at high risk (>20% chance) of having true ILD. On UVA, AI-ILD score, AI-ILD+ and MLD were significantly associated with the primary outcome of G2+ RP or mortality, as well as the secondary outcome of mortality. However, only MLD was significantly associated with the secondary outcome of G2+ RP. On MVA, both AI-ILD+ (OR 1.42, 95% CI 1.02-1.97, p = 0.04) and MLD (OR 1.13, 95% 1.05-1.21, p = 0.008) were significantly associated with G2+ RP or mortality. On Kaplan-Meier analysis, the median toxicity-free survival (TFS) time for AI-ILD+ and AI-ILD- patients were 1.7 and 3.4 years respectively, with a 2-year TFS of 48.3% vs. 59.3% (log-rank test: p = 0.02). There was no significant difference in rates of G2+ RP.

CONCLUSION

The AI-ILD algorithm can detect high risk patients with significantly decreased TFS following definitive treatment for NSCLC. AI-ILD classification was not associated with a significant difference in rates of RP when accounting for MLD. Future work will focus on improving the classification algorithm, expert radiologist validation of this dataset, and exploring reasons for the mortality difference in AI-ILD+ patients.

Prolonged QT dispersion in Subclinical Hypothyroid Females: A Study in University Teaching Hospital in Central Nepal

Background QT dispersion is a simple index derived from 12 lead ECG; its prolongation has been shown to be associated with increased arrhythmia risk. Increased cardiovascular risks, particularly occurrence of the malignant arrhythmias are a common finding in patients with subclinical hypothyroidism. This increased arrhythmia risk is found to be higher mainly in patients with TSH level more than 10 milli international unit per liter. Objective To assess QT dispersion among subclinical hypothyroid and euthyroid Nepalese females aged 20-59 years attending general practice out patient department of centrally located University Teaching Hospital from November 2016 to April 2017. Method Forty-three newly detected subclinical hypothyroid females and forty-one euthyroid females were enrolled. Resting electrocardiogram (ECG) was performed. QT dispersion was analyzed from ECG and corrected for heart rate using Framingham correction formula. Independent sample t-test was applied to compare mean QT dispersion between two groups. Pearson correlation test was used to examine the association between QT dispersion and TSH level. Result Mean QT dispersion for sub-clinical hypothyroid group was 75.35 ± 43.82 whereas mean QT dispersion for euthyroid group was 59.51 ± 22.13, with p value 0.041. A weak association between QT dispersion and TSH level was seen with correlation factor of 0.23. Conclusion The result showed prolongation of QT dispersion in sub-clinical hypothyroid group and weak positive correlation between TSH level and QT dispersion suggesting arrhythmia risk in subclinical hypothyroid females.

24-h personal monitoring of exposure to Power Frequency Magnetic Fields in adolescents - Results of a National Survey

OBJECTIVES

The aim of this exposure assessment study was to gain information about the exposure levels of adolescents in Israel to power frequency (50Hz) magnetic fields (MF) through personal monitoring, and to provide reliable data for national policy development.

METHODS

84 adolescents, 6-10th grade students, carried an EMDEX II meter attached to their body for 24h. The meter recorded the MF every 1.5s. The students documented their activities and microenvironments, such as apartment (awake or asleep), school, transportation, open public areas and other indoor environments.

RESULTS

The geometric mean (GM) of the daily time weighted average (TWA) of all the participants was 0.059 μT (STD = 1.83). This result is similar to those of personal exposure surveys conducted in the UK (GM 0.042-0.054μT), but lower than levels found in the US (GM 0.089 - 0.134μT). The arithmetic mean was 0.073μT, 23% higher than the GM. Fields were lowest at school (GM 0.033μT), and average outdoor exposures were higher than indoor ones. 3.6% of the participants were exposed to daily TWA above 0.2μT. The typical time spent above 0.2μT ranged from few minutes to few hours. The time spent above 0.4μT and 1μT were much shorter, around 1-15min and from few seconds to 2min, respectively. Momentary peaks ever recorded were in the range of 0.35-23.6μT CONCLUSIONS: Exposure of adolescents in Israel is similar to data reported in other countries, being below 0.1μT for the vast majority, with very few average exposures above 0.2μT. Analysis of the different microenvironments allows for a cost-effective and equitable policy development.

Yield of intensified tuberculosis case-finding activities using Xpert(®) MTB/RIF among risk groups in Nepal

SETTING

Twenty-two districts of Nepal, where intensified case-finding (ICF) activities for tuberculosis (TB) were implemented among risk groups under the TB REACH initiative in collaboration with the National TB Programme from July 2013 to November 2015.

OBJECTIVES

To assess the yield of TB screening using an algorithm with smear microscopy followed by Xpert(®) MTB/RIF.

DESIGN

A descriptive study using routinely collected data.

RESULTS

Of 145 679 individuals screened, 28 574 (19.6%) had presumptive TB; 1239 (4.3%) of these were diagnosed with TB and 1195 (96%) were initiated on anti-tuberculosis treatment. The yield of screening was highest among people living with the human immunodeficiency virus (PLHIV) (6.1%), followed by household contacts (3.5%) and urban slum dwellers (0.5%). Among other risk groups, such as prisoners, factory workers, refugees and individuals with diabetes, the yield was less than 0.5%. The number needed to screen to diagnose an active TB case was 17 for PLHIV, 29 for household contacts and 197 for urban slum dwellers. Of 11 525 patients from ICF and the routine programme, 112 (1%) were diagnosed with multidrug-resistant TB.

CONCLUSION

There was a substantial yield of TB cases among risk groups such as PLHIV and household contacts. Although the yield in urban slum dwellers was found to be moderate, some intervention should nonetheless be targeted because of the large population and poor access to care in this group.

Factors influencing parental consent for participation in clinical research involving their children in Egypt

Factors affecting parents' decision to involve their children in clinical research have not been studied in all cultural backgrounds. We aimed to explore the attitudes and beliefs influencing parents' decision to involve their children in clinical research in Mansoura, Egypt. Of 523 families approached, 357 filled the questionnaire. Only 98 (27.5%) parents consented to involve their child in clinical research. The children of consenters were significantly older than refusers: 8.6 (SD 7.2) versus 2.6 (SD 1.2) years. Factors favouring consent were: research of benefit to child (84.7%), enough explanation about the benefits (40.8%) and to learn more about child's condition (29.6%). Factors favouring refusal were: use of new drugs or vaccines (89.6%) and invasive procedures (84.2%). Parents' rate of consent was positively correlated with the research being non-invasive and the belief that research was of benefit to their child and negatively correlated with belief that refusal may negatively affect the care provided to their child.

Perception about the role of anesthesia and anesthesiologist among the paramedical staffs: perspective from a medical college in Nepal

BACKGROUND

Anesthesiologists and anesthesia has been considered behind the scene. The image and status of anesthesiologist in the eyes of the medical and lay communities has always been a problem.

OBJECTIVES

This study was designed to assess the knowledge about the role of anesthesiologist among the paramedical staffs at Kathmandu University Hospital.

METHODS

This prospective questionnaire based study was done at Kathmandu University School of Medical Science, Dhulikhel Hospital for 2nd January 2011 to 30th Jan 2011 among the paramedical staffs working in different department of the hospital.

RESULTS

There were 150 questionnaire distributed out of which 120 responded. Mean Age was 23.33 and most of the respondents were female with majority having education qualification equivalent to intermediate level. Only 49.20 said it to be a different specialty and 72.5% said anaesthesiologist work differently in the theatre where as 70% knew anaesthesiologist did something in the post-operative period too.

CONCLUSION

Anesthesiologists have duty to visit patients pre operatively and post operatively. The role inside the theatre and expanding role outside the theatre is poorly known. The awareness about the role of anesthesiologist in operation theatre, in intensive care unit, acute and chronic pain management and emergency care areas should be highlighted to all the staffs.

Enterprise Data Analysis and Visualization: An Interview Study

Organizations rely on data analysts to model customer engagement, streamline operations, improve production, inform business decisions, and combat fraud. Though numerous analysis and visualization tools have been built to improve the scale and efficiency at which analysts can work, there has been little research on how analysis takes place within the social and organizational context of companies. To better understand the enterprise analysts' ecosystem, we conducted semi-structured interviews with 35 data analysts from 25 organizations across a variety of sectors, including healthcare, retail, marketing and finance. Based on our interview data, we characterize the process of industrial data analysis and document how organizational features of an enterprise impact it. We describe recurring pain points, outstanding challenges, and barriers to adoption for visual analytic tools. Finally, we discuss design implications and opportunities for visual analysis research.

Comparison of sniffing position and simple head extension for visualization of glottis during direct laryngoscopy

Background The ability of good glottis visualization during direct laryngoscopy is major determinant of easy tracheal intubation. Sniffing position for laryngoscopy is considered as gold standard and ideal position. Several studies have questioned the validation of sniffing potion. Objectives This study aims to compare relative efficacy of sniffing position and simple head extension for visualization of glottis during direct laryngoscopy. Methods Four hundred patients undergoing elective surgery under general anesthesia requiring endotracheal intubation were randomized into two groups and study was concluded. Glottic visualization was assessed using modified Cormack and Lehane classification. After laryngoscopy, tracheal intubation was performed and intubation difficulty scale was noted. Results Both the groups were comparable regarding glottis visualization. Both the groups were comparable in demographic profiles. All the intubation difficulty scale variables were comparable in both the groups except N3. Total Intubation Difficulty Score was better in sniffing position than in simple head extension group. Conclusion Glottic visualization and intubation difficulty score was better in sniffing position as compared to simple head extension.http://dx.doi.org/10.3126/kumj.v9i1.6265 Kathmandu Univ Med J 2011;9(1):58-63

Detection of small pulmonary nodules in high-field MR at 3 T: evaluation of different pulse sequences using porcine lung explants

To evaluate two MR imaging sequences for the detection of artificial pulmonary nodules inside porcine lung explants. 67 agarose nodules ranging 3-20 mm were injected into ten porcine lungs within a dedicated chest phantom. The signal on T1-weighted images and radiopacity were adjusted by adding 0.125 mmol/l Gd-DTPA and 1.5 g/l of iodine. A T1-weighted three-dimensional gradient-echo (T1-3D-GRE; TR/TE:3.3/1.1 ms, slice:8 mm, flip-angle:10 degrees ) and a T2-weighted half-Fourier fast-spin echo sequence (T2-HF-FSE; TR/TE:2000/66 ms, slice:7 mm, flip-angle:90 degrees ) were applied in axial orientation using a 3-T system (Intera, Philips Medical Systems, Best, The Netherlands), followed by CT (16x0.5 mm) as reference. Nodule sizes and locations were assessed by three blinded observers. In nodules of >10 mm, sensitivity was 100% using 3D-GRE-MRI and 94% using the HF-FSE sequence. For nodules 6-10 mm, the sensitivity of MRI was lower than with CT (3D-GRE:92%; T2-HF-FSE:83%). In lesions smaller than 5 mm, the sensitivity declined to 80% (3D-GRE) and 53% (HF-FSE). Small lesion diameters were overestimated with both sequences, particularly with HF-FSE. This study confirms the feasibility of 3 T-MRI for lung nodule detection. In lesions greater than 5 mm, the sensitivity of the 3D-GRE sequence approximated CT (>90%), while sensitivity and PPV with the HF-FSE sequence were slightly inferior.

Influence of motor disorders on the visual perception of human movements in a case of peripheral dysgraphia

We report the case of a 71 year-old female patient (DC) with a left parietal lesion resulting in a peripheral dysgraphia essentially characterized by difficulties in letter sequences writing. The aim of our experiments was to analyze the influence of motor difficulties on the visual perception of both writing and reaching movements. Results showed a strong link between motor and perceptual performance. For reaching movements, performances in both production and perception tasks conform to the motor principles identified in healthy subjects (Fitts' law and motor anticipation). By contrast, for handwriting movements, DC's productions do not follow the motor principles usually observed in normal subjects (isochrony principle, motor anticipation) and in perception the same results were observed. The motor references used by DC in the visual perception of writing movement were not the laws of movement but rather her own way of writing. Taken together these data strongly suggest that motor competences is involved in the visual perception of human movements. They are discussed in the general framework of the simulation theory.

Detecting anticipatory events in handwriting movements

We investigated how visual processes exploit specific anticipatory movements observed in handwriting gestures. Previous research has shown that the kinematic information contained in the downstroke of an l is exploited to predict the identity of the forthcoming letter. Here, we determined the moment at which prediction takes place. Two between-letter effects were examined: changes in size (ll vs le) and changes in rotation direction (le vs ln). Results show that with only 75% of the l downstroke trajectory (or 60% of the downstroke time) subjects are already capable of predicting the identity of the letter following the l, that is well before the end of the downstroke. Analysis also reveals that identification takes place after the presentation of the movement acceleration phase. The visual perception of motor anticipation seems to involve the detection of motor events.

Perceptual anticipation in handwriting: the role of implicit motor competence

In two experiments, perceptual anticipation--that is, the observer's ability to predict the course of dynamic visual events--in the case of handwriting traces was investigated. Observers were shown the dynamic display of the middle letter l excerpted from two cursive trigrams (lll or lln) handwritten by one individual. The experimental factor was the distribution of the velocity along the trace, which was controlled by a single parameter, beta. Only for one value of this parameter (beta = 2/3) did the display comply with the two-thirds power law, which describes how tangential velocity depends on curvature in writing movements. The task was to indicate the trigram from which the trace was excerpted--that is, to guess the letter that followed the specific instance of the l that had been displayed. In Experiment 1, the no answer option was available. Experiment 2 adopted a forced-choice response rule. Responses were never reinforced. When beta = 2/3, the rate of correct guesses was high (Experiment 1, P¿correct¿ = .69; Experiment 2, P¿correct¿ = .78). The probability of a correct answer decreased significantly for both smaller and larger values of beta, with wrong answers becoming predominant at the extremes of the range of variation of this parameter. The results are consistent with the hypothesis that perceptual anticipation of human movements involves comparing the perceptual stimulus with an internal dynamic representation of the ongoing event.

Visual perception of motor anticipation in cursive handwriting: influence of spatial and movement information on the prediction of forthcoming letters

The execution of a graphemic sequence is constrained by spatial demands that result in fluctuations of letter shape and movement time. When producing two letters (ll, le, or ln) the movement time and the letter shape of the first letter depend on the execution constraints of the second one. The motor system thus anticipates the production of the forthcoming graphemic sequence during the production of the first letter. An experiment is reported the aim of which was to examine whether the visual system could exploit this anticipatory information to predict the identity of the letter following the l. Different ls belonging to ll, le, and ln were presented on a screen. Subjects had to predict to which couple of letters (ll, le, or ln) the presented l belonged to, by using information on the shape of the l and/or the movement that produced it. Results showed that the percentages of correct responses were higher in the conditions where the stimulus provided kinematic information than in the condition in which only spatial information was available. The ability to predict the forthcoming letter seems to be mediated by implicit knowledge on motor anticipation rules.

Effects of phonetic context on audio-visual intelligibility of French

Bimodal perception leads to better speech understanding than auditory perception alone. We evaluated the overall benefit of lip-reading on natural utterances of French produced by a single speaker. Eighteen French subjects with good audition and vision were administered a closed set identification test of VCVCV nonsense words consisting of three vowels [i, a, y] and six consonants [b, v, z, 3, R, l]. Stimuli were presented under both auditory and audio-visual conditions with white noise added at various signal-to-noise ratios. Identification scores were higher in the bimodal condition than in the auditory-alone condition, especially in situations where acoustic information was reduced. The auditory and audio-visual intelligibility of the three vowels [i, a, y] averaged over the six consonantal contexts was evaluated as well. Two different hierarchies of intelligibility were found. Auditorily, [a] was most intelligible, followed by [i] and then by [y]; whereas visually [y] was most intelligible, followed by [a] and [i]. We also quantified the contextual effects of the three vowels on the auditory and audio-visual intelligibility of the consonants. Both the auditory and the audio-visual intelligibility of surrounding consonants was highest in the [a] context, followed by the [i] context and lastly the [y] context.

A laminin-pepsin fragment with cell attachment and neurite outgrowth activity at distinct sites

Laminin is a large basement membrane glycoprotein which influences the behavior and morphology of a variety of cells. We have found that laminin and a pepsin fragment of laminin (P-lam) contain distinct sites for HT-1080 human fibrosarcoma cell attachment and for neurite outgrowth activity of PC12 and NG108-15 cell lines. Reduction and alkylation of laminin and P-lam fragment disulfide bonds, in the absence of denaturing agents, markedly reduced the cell attachment activity without reducing the neurite outgrowth response. The P-lam fragment (approximately 375 kDa) was found to contain part of the cross region of laminin and a portion of the long arm, on the basis of recognition by antisera against laminin synthetic peptides and fusion proteins. Modification of arginine residues by cyclohexanedione also had no effect on neurite outgrowth but reduced HT-1080 cell adhesion. Modification of lysine residues by succinic and citraconic anhydride, however, abolished laminin neurite outgrowth but not cell attachment activity. Neurite outgrowth activity was recovered by reversing the lysine modification. These data support the existence on laminin of separate sites for cell attachment and for neurite outgrowth.