International Journal of Laboratory Hematology最新文献

Introduction

Acute lymphoblastic leukemia (ALL) presents a formidable challenge in hematological malignancies, necessitating swift and precise diagnostic techniques for effective intervention. The conventional manual microscopy of blood smears, although widely practiced, suffers from significant limitations including labor-intensity and susceptibility to human error, particularly in distinguishing the subtle differences between normal and leukemic cells.

Methods

To overcome these limitations, our research introduces the ALLDet classifier, an innovative tool employing deep transfer learning for the automated analysis and categorization of ALL from White Blood Cell (WBC) nuclei images. Our investigation encompassed the evaluation of nine state-of-the-art pre-trained convolutional neural network (CNN) models, namely VGG16, VGG19, ResNet50, ResNet101, DenseNet121, DenseNet201, Xception, MobileNet, and EfficientNetB3. We augmented this approach by incorporating a sophisticated contour-based segmentation technique, derived from the Chan-Vese model, aimed at the meticulous segmentation of blast cell nuclei in blood smear images, thereby enhancing the accuracy of our analysis.

Results

The empirical assessment of these methodologies underscored the superior performance of the EfficientNetB3 model, which demonstrated exceptional metrics: a recall specificity of 98.5%, precision of 95.86%, F1-score of 97.16%, and an overall accuracy rate of 97.13%. The Chan-Vese model's adaptability to the irregular shapes of blast cells and its noise-resistant segmentation capability were key to capturing the complex morphological changes essential for accurate segmentation.

Conclusion

The combined application of the ALLDet classifier, powered by EfficientNetB3, with our advanced segmentation approach, emerges as a formidable advancement in the early detection and accurate diagnosis of ALL. This breakthrough not only signifies a pivotal leap in leukemia diagnostic methodologies but also holds the promise of significantly elevating the standards of patient care through the provision of timely and precise diagnoses. The implications of this study extend beyond immediate clinical utility, paving the way for future research to further refine and enhance the capabilities of artificial intelligence in medical diagnostics.

Introduction

The activity of direct oral anticoagulants (DOAC) is important in acute clinical situations. Recent studies have suggested a strong influence of DOAC on the diluted Russel's Viper Venom Time (dRVVT). Therefore, it may be a suitable screening parameter for antithrombotic plasma activity of different DOAC. This prospective study aims to evaluate the sensitivity and specificity of dRVVT to detect residual DOAC activity at recommended plasma level thresholds.

Methods

A total of 80 patients were recruited, with 20 each treated with one of the four approved DOAC (apixaban, edoxaban, rivaroxaban or dabigatran), respectively. Blood plasma was collected before (baseline), at plasma peak time, and 6 and 12 h after DOAC. DRVVT was measured using the screen (LA1) and confirm (LA2) assay for lupus anticoagulant and compared with DOAC plasma levels. A reference range was calculated based on the dRVVT values of 61 healthy blood donors.

Results

All DOAC significantly prolonged the dRVVT especially at higher DOAC plasma levels. The LA1 time ≥41 s had a sensitivity ≥98% to detect edoxaban, dabigatran and rivaroxaban plasma levels ≥30 ng/mL but it was only 87% for apixaban. Sensitivity was ≥98% for all DOAC with the LA2 assay ≥36 s. The negative predictive value of a DOAC plasma level <30 ng/mL and dRVVT LA2 <36 s was 99%.

Conclusions

The dRVVT confirm assay (LA2) reliably detects residual DOAC plasma levels ≥30 ng/mL and could be useful to rapidly rule out relevant DOAC activity in emergency situations and to guide treatment decisions.

Introduction

Recently, there has been an increasing interest to find a simple, low cost, widely available biomarker for outcome predictors in chronic obstructive pulmonary disease (COPD).

Methods

Absolute immature platelet count (AIPC), the percentage of AIPC to the total platelet count (immature platelet fraction [IPF%]), symptoms, spirometry results, age-dyspne-airflow obstruction index, and C-reactive protein tests of COPD patients and control group were recorded. Neutrophil/lymphocyte, monocyte/lymphocyte, and platelet/lymphocyte ratios and Charlson comorbidity index scores were calculated.

Results

One hundred and thirty-four COPD patients and 30 healthy control subjects were included in the study. Eighty-nine patients were in exacerbation (AECOPD) and 45 of them were in stable COPD period. There was a difference between IPF% values and AIPC of COPD group and control group (3.45 ± 2.41 vs. 2.04 ± 1.12, p = 0.01; 5.87 ± 2.45 vs. 5.20 ± 3.02, p = 0.01). A positive correlation was observed between IPF% with white blood cell count and neutrophil/lymphocyte ratio, platelet/lymphocyte ratio, monocyte/lymphocyte ratio in all patients (r = 0.352, p < 0.001; r = 0.399, p < 0.001; r = 0.186, p = 0.032; r = 0.200, p = 0.021) and AECOPD (r = 0.356, p < 0.001; r = 0.414, p < 0.001; r = 0.239, p = 0.025; r = 0.273, p = 0.010). At a cut-off of 3.4, IPF% showed the highest accuracy in identifying COPD (sensitivity: 80.3%, specificity: 82.5%) using receiver-operating characteristic analysis.

Conclusion

This is the first study to examine the relationship between AIPC, IPF%, and COPD. The higher IPF% values in COPD and the positive correlation between IPF% and other inflammatory markers are suggested that IPF may be an indicator of systemic inflammation in COPD.

Introduction

This observational study conducted across seven emergency care units compares the efficacy of four D-dimer detection methods, namely HemosIL D-dimer HS (HS), HemosIL D-dimer HS-500 (HS-500), VIDAS D-dimer (VIDAS), and HemosIL AcuStar D-dimer (ACUSTAR). The primary focus is on patients with a clinical suspicion of deep venous thrombosis (DVT) or pulmonary embolism (PE).

Methods

A total of 149 samples were collected from patients with suspected DVT or PE. The confirmation of DVT/PE was based on calf ultrasound or computed tomography-Angiography. Direct comparisons were made between the different detection methods, considering both their analytical performance and clinical utility. Additionally, the impact of an age-adjusted cut-off on the diagnostic accuracy of each method was assessed.

Results

The results revealed comparable negative predictive value, sensitivity, and specificity across the methods, with a notable exception of increased specificity for HS compared with HS-500 (50.8% vs. 41.5%, p = 0.03). Further analysis incorporating an age-adjusted cut-off demonstrated a significant improvement in specificity for HS. When using the age-adjusted cut-off, HS exhibited a substantial increase in specificity compared with HS-500 (63.1% vs. 49.2%, p = 0.004) and demonstrated significantly higher specificity compared with VIDAS (63.1% vs. 53.8%, p = 0.04).

Conclusion

The study emphasizes the nonuniversal effect of an age-adjusted cut-off and discusses the potential necessity for different cut-off values, particularly in the case of HS-500. These findings contribute to the understanding of D-dimer detection methods in the context of DVT and PE, providing insights into their relative performances and the potential optimization through age-adjusted cut-offs.

Background

In this study, we combined two techniques, ultrasound-guided needle biopsy and flow cytometry (FCM), to explore their value in patients with enlarged lymph nodes.

Methods

We compared the results of 198 needle biopsies on FCM and pathology. Forty-two were done by (fine needle aspiration, FNA), and the remaining 156 with (core needle biopsy, CNB), in 36 of 156 patients, a FNA was performed in the same lymph node after completion of the CNB. Except for five types of pathological entities, the rest were differentiated only detected or undetected tumours as the outcome distinction.

Results

Among the 198 needle biopsies, 13 were inadequate specimens, while the remaining 185 had pathological findings, including 47 benign and 138 neoplastic findings. Thirty-six patients underwent puncture with both FNA and CNB, both needles produced identical results by FCM, but more cells were obtained by FNA. Among the pathologically positive results, there were 23 missed diagnoses in FCM, in contrast, evidence of tumours was observed in the FCM images of 15 needle biopsies that reported benign or findings that were inconsistent with pathology, and the final diagnosis was consistent with the FCM in 10 cases. FCM detected haematolymphoid tumours with a sensitivity of 87.8% and a specificity of 91.9%.

Conclusion

The combination of FCM and ultrasound-guided lymph node needle biopsy can quickly provide guidance for clinical decision-making. We recommend that all lymph node needle biopsies be sent for FCM, the specimen can be obtained by the last puncture with FNA.

Introduction

Candidemia can be a significant cause of death in immunosuppressed or debilitated patients particularly. Abnormalities of the instrumental cytograms of some hematological analyzers, such as Mindray BC-6800Plus, can be related to circulating Candida. We studied the possible diagnostic usefulness of this information.

Methods

A fungal bloodstream infection has been simulated by adding aliquots of Candida albicans, Candida parapsilosis, and Candida glabrata to 75 leftovers and anonymized peripheral blood samples. Cytographic abnormalities like those of experimental samples were used to select patients with possible fungemia. The microscopic review of peripheral blood smears constituted the confirmatory method.

Results

In all experimental samples, the various Candida types caused pseudo-NRBC and morphological abnormalities of WNB and DIFF cytograms. Circulating blastospores, free or engulfed by neutrophils, were the microscopic findings in the peripheral blood smears.

In the clinical verification, 72 patients were recruited based on the presence of an evocative cluster in the WNB and DIFF cytograms. The microscopic review of 39 out of 72 samples was positive for NRBC. According to blood cultures, light microscopy revealed fungal forms of several Candida or non-Candida types in the remaining 33 samples. Nine of these cases were not yet known to suffer from bloodstream infection.

Conclusions

Although further confirmatory clinical studies are required for these diagnostic abilities, the BC 6800Plus cytographic abnormalities related to fungemia have proven helpful in rapidly monitoring persistent fungemia in already diagnosed patients. In unknown or undiagnosed cases, they could be the trigger point for the subsequent diagnostic-therapeutic pathway.

Introduction

Leukemic stem cells (LSCs) are the transcriptionally low/silent cells which are resistant to the tyrosine kinase inhibitor. These have been found to play a pivotal role in disease relapse in chronic myeloid leukemia (CML) cases. The present study evaluated the correlation of absolute CML-LSC count in the peripheral blood (PB) at diagnosis and achievement of major molecular response (MMR) at 12 months in patients of CML-CP.

Methods

This was a prospective, observational, non-interventional single center study including newly diagnosed adult (>18 yrs) CML-CP patients. Absolute CD26 + CML-LSC quantification was done by multiparametric flow cytometry. Patients were treated with Imatinib treatment and subsequently monitored at 3-month intervals for BCR::ABL transcript levels. MMR was defined as a BCR::ABL1 transcript level of less than 0.1% on international scale.

Results

A total of 89 patients were enrolled in the study out of which 40.5% achieved MMR at 12 months. There was a significant difference in the median absolute CML-LSC count of the patients who achieved MMR at 12 months as compared to those who did not (58.5 vs 368.1 cells/μL; p value <0.001). Using a ROC analysis, a count of <165.69 CML LSC/μL was identified to have a sensitivity of 83.8% and specificity of 72.4%, in predicting the MMR at 12 months.

Conclusion

Absolute CML-LSC count at diagnosis in the PB predicts the MMR achievement at 12 months. An absolute count of less than 165 cells/μL is highly predictive of achieving MMR at 12 months.