Enhancing histopathological image classification of invasive ductal carcinoma using hybrid harmonization techniques

This study aims to develop a robust pipeline for classifying invasive ductal carcinomas and benign tumors in histopathological images, addressing variability within and between centers. We specifically tackle the challenge of detecting atypical data and variability between common clusters within the same database. Our feature engineering-based pipeline comprises a feature extraction step, followed by multiple harmonization techniques to rectify intra- and inter-center batch effects resulting from image acquisition variability and diverse patient clinical characteristics. These harmonization steps facilitate the construction of more robust and efficient models. We assess the proposed pipeline's performance on two public breast cancer databases, BreaKHIS and IDCDB, utilizing recall, precision, and accuracy metrics. Our pipeline outperforms recent models, achieving 90-95% accuracy in classifying benign and malignant tumors. We demonstrate the advantage of harmonization for classifying patches from different databases. Our top model scored 94.7% for IDCDB and 95.2% for BreaKHis, surpassing existing feature engineering-based models (92.1% for IDCDB and 87.7% for BreaKHIS) and attaining comparable performance to deep learning models. The proposed feature-engineering-based pipeline effectively classifies malignant and benign tumors while addressing variability within and between centers through the incorporation of various harmonization techniques. Our findings reveal that harmonizing variabilities between patches from different batches directly impacts the learning and testing performance of classification models. This pipeline has the potential to enhance breast cancer diagnosis and treatment and may be applicable to other diseases.

Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: investigating ComBat strategies, sub-volume characterization, and automatic segmentation

PURPOSE

This study aimed to investigate the impact of several ComBat harmonization strategies, intra-tumoral sub-volume characterization, and automatic segmentations for progression-free survival (PFS) prediction through radiomics modeling for patients with head and neck cancer (HNC) in PET/CT images.

METHODS

The HECKTOR MICCAI 2021 challenge set containing PET/CT images and clinical data of 325 oropharynx HNC patients was exploited. A total of 346 IBSI-compliant radiomic features were extracted for each patient's primary tumor volume defined by the reference manual contours. Modeling relied on least absolute shrinkage Cox regression (Lasso-Cox) for feature selection (FS) and Cox proportional-hazards (CoxPH) models were built to predict PFS. Within this methodological framework, 8 different strategies for ComBat harmonization were compared, including before or after FS, in feature groups separately or all features directly, and with center or clustering-determined labels. Features extracted from tumor sub-volume clustering were also investigated for their prognostic additional value. Finally, 3 automatic segmentations (2 threshold-based and a 3D U-Net) were also compared. All results were evaluated with the concordance index (C-index).

RESULTS

Radiomics features without harmonization, combined with clinical factors, led to models with C-index values of 0.69 in the testing set. The best version of ComBat harmonization, i.e., after FS, for feature groups separately and relying on clustering-determined labels, achieved a C-index of 0.71. The use of features extracted from tumor sub-volumes further improved the C-index to 0.72. Models that relied on the automatic segmentations yielded close but slightly lower prognostic performance (0.67-0.70) compared to reference contours.

CONCLUSION

A standard radiomics pipeline allowed for prediction of PFS in a multicenter HNC cohort. Applying a specific strategy of ComBat harmonization improved the performance. The extraction of intra-tumoral sub-volume features and automatic segmentation could contribute to the improvement and automation of prognosis modeling, respectively.

Lung and colon cancer classification using medical imaging: a feature engineering approach

Lung and colon cancers lead to a significant portion of deaths. Their simultaneous occurrence is uncommon, however, in the absence of early diagnosis, the metastasis of cancer cells is very high between these two organs. Currently, histopathological diagnosis and appropriate treatment are the only way to improve the chances of survival and reduce cancer mortality. Using artificial intelligence in the histopathological diagnosis of colon and lung cancer can provide significant help to specialists in identifying cases of colon and lung cancers with less effort, time and cost. The objective of this study is to set up a computer-aided diagnostic system that can accurately classify five types of colon and lung tissues (two classes for colon cancer and three classes for lung cancer) by analyzing their histopathological images. Using machine learning, features engineering and image processing techniques, the six models XGBoost, SVM, RF, LDA, MLP and LightGBM were used to perform the classification of histopathological images of lung and colon cancers that were acquired from the LC25000 dataset. The main advantage of using machine learning models is that they allow a better interpretability of the classification model since they are based on feature engineering; however, deep learning models are black box networks whose working is very difficult to understand due to the complex network design. The acquired experimental results show that machine learning models give satisfactory results and are very precise in identifying classes of lung and colon cancer subtypes. The XGBoost model gave the best performance with an accuracy of 99% and a F1-score of 98.8%. The implementation and the development of this model will help healthcare specialists identify types of colon and lung cancers. The code will be available upon request.

Investigating the impact of remediation efforts on Hedistediversicolor in the Seine estuary using multiple levels of biological organization

The Seine Estuary is historically one of the most contaminated estuaries in Europe. In 2002 the estuary underwent major redevelopment which emphasized environmental remediation to measure biological effects. The present study investigated the health status of the sentinel species Hediste diversicolor in the Seine estuary by comparing data from a sampling campaign immediately after the Seine underwent reconstruction and 10 years afterward. Both studies implemented multiple levels of biological organization ranging from enzymatic biomarkers to population density. Integrative modeling was used to establish a holistic status assessment for H. diversicolor in the Seine. H. diversicolor populations in the Seine estuary were impacted in both campaigns suggesting no significant improvements over the 10-year period. This interpretation however can be improved by increasing the number of reference sites used to establish the baseline natural variation, integrating additional environmental parameters into the model and measuring additional biological endpoints, such as reproductive status.

Baseline levels of biochemical biomarkers in the endobenthic ragworm Hediste diversicolor as useful tools in biological monitoring of estuaries under anthropogenic pressure

Identification of contamination in estuarine ecosystems that are impacted by anthropogenic pressures, such as the Seine estuary, is difficult to determine without considering the role environmental variation plays on the end points selected. Currently, there is interest in identifying methods in which the influence of confounding factors can be described and accounted for. In this context, the aim of this study was to define a baseline assessment criteria (BAC) for enzymatic biomarkers in ragworms (Hediste diversicolor) collected in a reference site (Authie). The model took into consideration the weight, temperature and salinity of the site. Values collected in the Seine estuary were analyzed with the model to determine if differences between the sites could potentially be due to contamination or were explained by environmental variation. In general, biomarker responses from the Seine estuary fell within the range of BAC, suggesting that environmental variation could explain some of the results.

[Social Autonomy Scale. First validation data]

The Assessment of Social Self-Sufficiency (ASSS) is a scale of hetero-assessment, made up of seventeen items and five subsets, devised to investigate the self-sufficiency level of the persons who suffer from chronic psychiatric trouble. We know indeed that, for these patients, the success of their rehabilitation plans depends on their ability to face up to tangible problems in everyday life. This publication expounds the first work validating the ASSS. The "intercotators" accuracy of the work, its sensitiveness, its concurrent validity investigated with regard to the Global Assessment of Functioning Scale (GAF of the DSM III-R) as well as the inner contents seem to be very satisfactory.

Squamous cell carcinoma in epidermolysis bullosa. Treatment with systemic chemotherapy

Dystrophic epidermolysis bullosa is associated with a high incidence of cutaneous squamous cell carcinoma. Despite aggressive surgical treatment, metastases occur frequently, and survival is generally poor. Chemotherapy for advanced disease has usually been avoided because of the potential for severe cutaneous toxicity. Two patients with autosomal recessive dystrophic epidermolysis bullosa and advanced squamous cell carcinoma are described. Both received cisplatin-based systemic chemotherapy without significant toxicity.

Necrotizing arteritis associated with blastomycosis

We have described a patient with cavitary pulmonary fungal infection, probably blastomycosis, with necrotizing arteritis of the skin. The pulmonary and skin lesions resolved with amphotericin B therapy.

Precedents for meaningful recovery during treatment in a medical intensive care unit. Outcome in patients with hematologic malignancy

The medical records of 77 patients with hematologic malignancy who were admitted to a medical intensive care unit over a 21-month period were reviewed. The overall hospital mortality rate was 80 percent. Sixteen patients (21 percent) were discharged from the intensive care unit but eventually died in the hospital. The cause of death was the result of a new problem in only three of these 16 patients. Hypotension (shock) and acute respiratory failure were the reasons prompting admission to the intensive care unit in 75 percent, but death in the intensive care unit was almost always the result of intractable hypotension rather than refractory hypoxemia. Only four of 52 patients who required mechanical ventilation left the hospital. In all four, the duration of ventilatory support was less than five days and the cause of respiratory failure was noninfectious in nature. Factors such as congestive heart failure, leukopenia, and abnormalities in mental status modified the hospital course, but did not alter outcome once prolonged mechanical ventilation became necessary. The data suggest that once acute respiratory failure develops in patients with lymphoma or leukemia, presumably as a result of infection, and mechanical ventilation for more than a relatively brief period is required, the prognosis is uniformly grim. Decisions to limit aggressive therapies is subsets of intensive care patients such as these should be aided by data that show a lack of precedent for meaningful recovery.

Unilateral lung function. Comparison of the lateral position test with radionuclide ventilation-perfusion studies

Twenty-seven patients underwent studies of unilateral lung function by the lateral-position test (LPT) and by computer-analyzed radionuclide imaging of ventilation and perfusion. The patients were divided into two groups, symmetric or asymmetric, on the basis of the physical examination of the chest and the chest radiograph. In patients with symmetry, the estimate of unilateral lung function by the LPT and isotopic estimates for unilateral lung volume, unilateral distribution of tidal volume, and unilateral perfusion, agreed within 2 percent, 4 percent, and 3 percent, respectively. In patients with asymmetry, the differences were 9 percent, 8 percent, and 13 percent. In settings of marked unilateral ventilation-perfusion imbalance, the LPT primarily reflected ventilation. Prediction of unilateral ventilatory function based upon the LPT and spirometric measurements agreed closely with unilateral ventilation determined isotopically by 133xenon, even in the presence of chronic obstructive lung disease. Our results confirm that the LPT provides valid information about unilateral lung function.