Pakistan Journal of Medical & Health Sciences

Bridging the Peripheral Divide: The Emerging Horizon of AI-Assisted Peripheral Bronchoscopy

Akhtar Ali — Sat, 10 Oct 2026 00:00:00 +0400

Lung cancer screening using low-dose computed tomography (LDCT) has significantly helped to detect numerous small and indeterminate peripheral pulmonary lesions (PPLs). Despite early radiological detection being an opportunity offering a chance to cure the disease, the definite coronation of tissues using small and distal lesions is still a primary issue in interventional pulmonology. Traditional forms of flexible bronchoscopy and early navigational methods might be restricted by CT-to-body divergence (CTBD), respiratory motion, anatomical fluctuation, operator dependence and the inability to keep the catheter intact on the target lesion. This means the vicinity around a lesion is not always reached to attain a diagnostic tissue acquisition^1-3.

Artificial intelligence (AI) is also sought out as an adjunctive technology to overcome these limitations. The combination of the so-called deep learning, computer vision, image analysis, and computational pathology with intelligent navigation can turn peripheral bronchoscopy into a more accurate and standardized diagnostic process^2-4.

The distal bronchial airways are difficult to visualize due to the tapering airway size and the low optical field. Newer ultra-thin bronchoscopic systems and video probes can allow visualization to a more distal airway and potentially using this system to examine small PPLs directly. Computer vision AI algorithms can also be used to aid in the interpretation of bronchoscopic images, by detecting minute morphological changes that could be related to malignancy^2-5.

Radial-probe endobronchial ultrasound (r-EBUS) is another procedure that is being applied with AI to localize peripheral lesions with common usage. The analysis of the r-EBUS patterns may also be difficult due to the fact that the benign and the benign lesions may display confused or mixed results. Transfer-learning models and CNN-based models demonstrated good diagnostic accuracy in the distinction between malignant and benign peripheral lesions. Lower level CNN-Transformer structures can also be enhanced to improve image classification and segmentation by incorporating both local features detailed and documenting contextual features^3-6.

The AI application is not just limited to localizing lesions but also to the appraisal of tissue gained through the bronchoscopic examination. Pathomics Computational pathology involves artificial intelligence applied to obtain quantitative morphology in histopathology images. Deep learning models have the ability to detect cellular and architectural features which can be hard to detect on a regular basis using traditional microscopic evaluation. Combination of deep pathomic aspects with clinical and radiological variables could enhance the delimiting of peripheral pulmonary lesions. These models have shown capabilities to exhibit good diagnostic performance and can also help in further revealing fine malignant features in specimens that can otherwise be challenging to determine^3,5-7.

Other applications that have been brought by AI include rapid on-site evaluation (AI-ROSE). In bronchoscopic biopsy, specimen adequacy can be immediately evaluated and specimen additional needs determined. Nevertheless, there is a lack of on-site cytopathologists in most institutions. Automated evaluation of adequacy of the specimen at the cytological imaging phase using deep learning algorithms and help distinguish between malignant and benign or inflammatory tissue can be applied to the cytological imaging of the sample. Such method can minimise poor sampling and even lessen the use of repeated invasive procedures^2,4-5.

The most promising clinical use of AI could be the combination with robotic-assisted bronchoscopy (RAB) and enhanced navigation systems. Robotic systems offer better catheter stability and articulation and distal airway access. These capabilities can be enhanced with the help of AI in the field of automated segmentation of the airways, three-dimensional navigation of the pathways, finding the target and catheter positioning. One significant issue with peripheral bronchoscopy is the CT-to-body divergence, where the position of the lesion at the time of the procedure is not the same as it is on the CT image before the procedure due to respiratory motion, deformation of the lung, atelectasis and changes in patient positioning. AI-aided navigation could be integrated with real-time imaging examinations like cone-beam CT (CBCT) and digital tomosynthesis to find these anatomical discrepancies and enhance distinctions in accuracy of targeting^3,5,7-9.

The advent of AI, robotic bronchoscopy, and real-time imaging is hence a significant advance in the direction of precise and controlled biopsy of peripheral lesions. However, the reported diagnostic yields depend on the lesion size, location, bronchus sign, sampling method, experience of the operator and strategy of imaging and need additional prospective assessment.

AI-assisted peripheral bronchoscopy is a promising new frontier of interventional pulmonology. Its uses include bronchoscopic interpretation of images, r-EBUS evaluation, lesion navigation, robotic bronchoscopy, computational pathology, and real-time specimen evaluation. AI could enhance the capabilities of conventional peripheral bronchoscopy that have a number of shortcomings by enhancing visual interpretation capability, localizing targets better, and assisting in the evaluation of tissue adequacy.

Diaphragmatic Thickening Fraction as a Predictor of Early Treatment Failure in Acute Exacerbation of COPD

Sun, 30 Aug 2026 00:00:00 +0400

Background: If acute exacerbation of chronic obstructive pulmonary disease is not prevented, it can develop into hypercapnic respiratory failure and noninvasive ventilation failure. Bedside ultrasonography of diaphragm thickening fraction could be used as an early predictor of respiratory muscle dysfunction.

Objective: To determine whether diaphragmatic thickening fraction predicts early treatment failure in patients with acute exacerbation of chronic obstructive pulmonary disease receiving noninvasive ventilation.

Methods: This was a prospective observational cohort study of 200 adults who were admitted in Jinnah Hospital Lahore from March 2024 to March 2026. Diaphragm thickness was measured at end-expiration and peak inspiration at baseline and two hours after starting noninvasive ventilation. The thickness of the diaphragm was measured and the average difference between the inspiratory and expiratory thickness was expressed as a percent of the inspiratory thickness, as the diaphragmatic thickness fraction. Failure of treatment was defined as endotracheal intubation or death in less than 72 hours.

Results: Early treatment failure occurred in 52 patients (26.0%). The baseline diaphragmatic thickening fraction was lower in the patients who failed treatment, as compared to those who were successfully treated (18.5 ± 6.3% vs 28.2 ± 7.6%, p<0.001). The difference was greater after two hours (17.9 ± 6.2% versus 32.4 ± 7.5%; p<0.001). A two hour threshold of ≤23% detected failure with sensitivity of 86.5%, specificity of 82.4% and area under the curve of 0.91. Diaphragmatic thickening fraction ≤23% was the best independent predictor of failure (adjusted odds ratio 9.72; 95% confidence interval 4.16–22.71; p<0.001) after adjustment. It has a high negative predictive value, potentially helping to identify those patients who are likely to continue to receive noninvasive ventilation successfully.

Conclusion: The 2-hour diaphragmatic thickening fraction was an independent predictor of early treatment failure and might help guide timely escalation of respiratory care.

Keywords: acute exacerbation of COPD; diaphragmatic thickening fraction; diaphragm ultrasonography; noninvasive ventilation; respiratory failure; treatment failure

Ultrasound-Derived Lung Aeration Score for Predicting Recovery in Severe Community-Acquired Pneumonia

Akhtar Ali, Hina Awan, Muntaqa Butt, Noor E Seher, Hamza Akram, Muhammad Kashif — Sun, 30 Aug 2026 00:00:00 +0400

Background: Severe community-acquired pneumonia is linked to respiratory failure, long hospital intensive care and significant mortality. Patients not likely to recover can be identified early to provide better surveillance and management. The present study assessed the utility of serial ultrasound-derived lung aeration scores for early clinical recovery prediction.

Methods: This was a prospective observational cohort study of 80 adults with severe community-acquired pneumonia (CAP) from June 2025 to March 2026 at Shalamar Hospital Lahore. Lung ultrasound was done within 24 hours of admission, and assessed after 72 hours with a standardized 12-zone protocol. The scores for each region ranged from 0 to 3 and the scores for all regions were summed, resulting in a total score of 0–36. The primary outcome was clinical recovery at day 7. Receiver operating characteristic analysis and multivariable logistic regression were used to assess predictive performance.

Results: Clinical recovery occurred in 52 patients (65.0%), whereas 28 patients (35.0%) failed to recover. Non-recovered patients had higher admission scores than recovered patients (22.4 ± 4.6 vs 16.8 ± 4.3; p<0.001) and markedly higher day-3 scores (20.1 ± 5.0 vs 10.7 ± 4.6; p<0.001). A day-3 score of >16 had a sensitivity of 82.1%, a specificity of 84.6% and an area under the curve of 0.91, predicting non-recovery. For every 1-point improvement on day-3 score, there was a 31% increase in the odds of non-recovery. It can be used to add to traditional clinical and oxygenation severity assessment.

Conclusion: Loss of lung aeration at 72 hours was highly associated with delayed clinical recovery. Serial lung ultrasound could prove to be a useful bedside tool for early risk-stratification in severe community-acquired pneumonia.

Clinical Frailty–Inflammation Phenotypes and Their Association with Treatment Failure in Older Medical Inpatients

Sun, 30 Aug 2026 00:00:00 +0400

Background: Frailty and systemic inflammation could have a combined effect on recovery from acute illness in older people but their combined prognostic significance in general medical inpatients has not been fully characterized.

Objective: To determine the association between admission frailty–inflammation phenotypes and treatment failure among older medical inpatients.

Methods: This was a prospective observational cohort study of 100 patients ≥65 years of age during a 12-month period (January 2024 to January 2025) who were admitted to general medical wards at Sheikh Zayed Hospital Rahim Yar Khan. Frailty was determined by Clinical Frailty Scale (CFS); scores of 5–8 indicate frailty. Patients were divided into 4 groups: Robust–Low inflammation, Robust–High inflammation, Frail–Low inflammation, and Frail–High inflammation, according to the inflammation category. Treatment failure was defined as lack of clinical stability at day 5, escalation of major treatment, transfer to intensive care, or death in the hospital.

Results: There were 47 male and 53 female patients with the mean age 75.6 ± 7.5 years. 30 patients had treatment failure. Rates were 10.7%, 25.0%, 27.8%, and 50.0% across the four phenotypes, respectively (p=0.008). The frail–high inflammation phenotype was still associated with treatment failure (adjusted OR 5.72, 95% CI, 1.31–25.03; p=0.020) after adjustment. This group was also the group with the longest hospital stay, and the greatest incidence of functional decline.

Conclusions: Frailty and high inflammation were found to be most important in predicting treatment failure in older patients admitted to hospital. Clinical Frailty Scale and C-reactive protein provide a combined assessment which might be useful in the early risk stratification and individual inpatient management and to allow earlier multidisciplinary interventions during acute hospitalisation.

Urinary Chloride-to-Creatinine Ratio as an Early Predictor of Diuretic Resistance in Acute Heart Failure

Sun, 30 Aug 2026 00:00:00 +0400

Background: Diuretic resistance is an important issue in the management of acute heart failure, as it is associated with slow decongestive response and high likelihood of ongoing fluid retention, renal failure and longer hospital stays. The aim of this study was to assess the usefulness of the urinary chloride to creatinine ratio as an early predictor for inadequate diuretic response.

Methods: This is a prospective, observational study of 110 adults hospitalized at Mayo Hospital Lahore with acute heart failure from April 2025 to April 2026. Spot urine was done two hours following the first dose of intravenous Furosemide. Diuretic resistance was considered to be low urine output on 24-hour basis, poor weight loss and intensification of diuretic therapy. ROC analysis and multivariable logistic regression were undertaken.

Results: Diuretic resistance occurred in 35 patients (31.8%). The median urinary chloride to creatinine ratio was significantly lower in resistant compared to responsive patients (56 [38-72] vs 118 [86-156] mmol/g, p<0.001). A cut-off of < 72 mmol/g was shown to have 82.9% sensitivity, 78.7% specificity, and an area under the curve of 0.86 for predicting resistance. A ratio <72 mmol/g was independently associated with resistance after the adjustment (adjusted odds ratio 6.18; 95% confidence interval 2.43–15.69; p<0.001). Resistant patients were also less likely to have high urine output, less net fluid loss, more worsening renal function, persistent congestion, and shorter hospital stay. After adjustment for kidney function, systolic blood pressure, chronic use of a loop diuretic, serum chloride and sex, the association was still present, suggesting that it is independently predictive in routine clinical practice settings.

Conclusion: Urinary chloride to creatinine ratio is a fast, simple, and clinically helpful method of rapid assessment of early diuretic resistance in acute heart failure.

Admission Lactate-to-Albumin Ratio as a Predictor of Severe Maternal Outcomes in Postpartum Haemorrhage

Sun, 30 Aug 2026 00:00:00 +0400

Background: Postpartum haemorrhage is an obstetric emergency which may quickly escalate to shock, coagulopathy, organ failure and death. Conventional vital signs and haemoglobin may underestimate the severity of blood-loss early on, and the identification of women at risk is difficult.

Objectives: To assess the predictive value of admission lactate/albumin ratio (LAR) in women with postpartum haemorrhage (PPH) on severe maternal outcomes.

Methods: This prospective observational cohort study involved 70 women at Lady Willingdon Hospital, who had primary postpartum haemorrhage from April 2025 to January 2026. Venous lactate and serum albumin were obtained within 30 minutes of recognition of haemorrhage and lactate to albumin ratio (L:ALB) was calculated. Severe maternal outcome defined as either maternal death, peripartum hysterectomy, transfusion of ≥5 PRBC units, vasopressor support, disseminated intravascular coagulation, invasive mechanical ventilation, cardiac arrest or renal replacement therapy. ROC and logistic regression analyses were used.

Results: There were 18 women (25.7%) with severe maternal outcomes. The median lactate-to-albumin ratio was significantly higher in women with severe outcome than in women without severe outcome (1.72 versus 0.66; p<0.001). The ratio demonstrated good discriminatory ability with an area under the curve of 0.88. With a threshold value of 0.95, 83.3% sensitivity, 78.8% specificity, 57.7% PPV and 93.2% NPV were obtained. The ratio, after taking into account estimated blood loss, was independently associated with severe maternal outcomes at a ratio of 0.95 or less (adjusted odds ratio 6.74; 95% confidence interval 1.72 – 26.46, p=0.006).

Conclusions: Admission lactate/albumin ratio could be a tool to identify women at risk of complications after PPH.