Image analysis on corneal opacity: A novel method to estimate postmortem interval in rabbits

Application of artificial intelligence and machine learning technology for the prediction of postmortem interval: A systematic review of preclinical and clinical studies

BACKGROUND /PURPOSE

Establishing an accurate postmortem interval (PMI) is exceptionally crucial in forensic investigation. Artificial intelligence (AI) and Machine learning (ML) models are widely employed in forensic practice. ML is a part of AI, both terms are highly associated and sometimes used interchangeably. This systematic review aims to evaluate the application and performance of AI technology for the prediction of PMI.

METHODS

Systematic literature search across different electronic databases using PubMed/Google Scholar/EMBASE/Scopus/CINAHL/Web of Science/Cochrane library was conducted from inception to 3 December 2021 for preclinical and clinical studies reported ML models for PMI estimation.

RESULTS

We identified 18 studies (12 preclinical and 06 clinical) that met the inclusion criteria in the qualitative analysis. Most of the studies employed supervised learning (N = 15), and others employed unsupervised learning (N = 3). Due to the heterogeneity of the samples, quantitative analysis was not performed.

CONCLUSION

In this systematic review, we discussed the performance of AI-based automated systems in PMI estimation. ML models have demonstrated accuracy and precision and the ability to overcome human errors and bias. However, the research is limited, conducted in primarily small, selected human populations. In addition, we suggest further research in larger population-based studies is needed to fully understand the extent of integrated ML models.

Corneal absorption spectra in the deep UV range

SIGNIFICANCE

Refractive surgery in ophthalmology uses pulsed lasers at 193, 210, or 213 nm. The reason is that most molecular constituents of cornea absorb strongly in this wavelength range. Precise refractive surgery via ablation requires an accurate knowledge of the absorption coefficient at the relevant wavelengths. Yet, the absorption coefficients of corneal tissue reported in literature vary by almost an order of magnitude; moreover, they were measured mostly at the wavelengths mentioned earlier.

AIM

By measuring the corneal absorption coefficient of intact eyeballs stored at different environmental conditions, prepared by following different procedures, and as a function of postmortem time, we determine the absorption coefficient for the entire wavelength range between 185 and 250 nm for as close as possible to in-vivo conditions.

APPROACH

We use a specially designed UV ellipsometer to measure refractive index and absorption coefficient. Specifically, we investigate the temporal evolution of refractive index and absorption coefficient after enucleation of the eyeballs under different environmental conditions and preparation procedures.

RESULTS

Our measurements provide accurate values for refractive index as well as absorption coefficient of cornea in the wavelength range between 185 and 250 nm. We find that the absorption coefficient decreases with time and that neither storage conditions nor preparation procedures but a continuous degeneration of the cornea is responsible for the observed time evolution. We use the measured time evolution to extrapolate refractive index and absorption coefficient to in-vivo conditions.

CONCLUSION

Our measurements of the close to in-vivo absorption coefficient of cornea between 185 and 250 nm allow for a better understanding and modeling of refractive cornea surgery, also at other than the three commonly used wavelengths. In the future, this may be relevant when new pulsed laser sources with other wavelengths become available.

The Eye in Forensic Medicine: A Narrative Review

ABSTRACT

The eye, with its distinctive anatomy, not only reflects a wide variety of diseases in life but also undergoes a myriad of post-mortem changes. Consequently, the eye has long been an area of interest in forensic science, primarily for the estimation of post-mortem interval and therefore the time of death and also for assistance in ascertaining the cause of death. There has been significant progress in the knowledge of ophthalmic forensic science using new technologies which have allowed further possibilities to arise where understanding of this field can assist the forensic pathologist. This review aims to highlight the current knowledge which exists in this field and also to identify important avenues for further investigation. Post-mortem changes of the eye along with its current applications and challenges will be discussed. These include the important areas of post-mortem iris biometrics, pupil size correlation with post-mortem interval, use of point-of-care technology on vitreous humor, and the use of ophthalmic imaging in pediatric abusive head trauma.

A Cross-Sectional Study of Time Since Death From Image Analysis of Corneal Opacity

Estimation of time since death (TSD) is an important aspect of forensic medicine. Methods used so far are subjective and have human errors. Corneal opacity images using software to analyze the red, green, and blue (RGB) components of corneal color against the TSD may prove to be an objective method. This study aimed to estimate TSD from image analysis of corneal opacity from the cornea of deceased individuals brought in for medicolegal autopsy to study the factors affecting corneal opacity and to formulate a predictive equation for the estimation of TSD. This was a cross-sectional study conducted at the department of forensic medicine and toxicology of a tertiary care medical institute over two years. The study group included cadavers brought in for autopsy where the TSD was known from hospital records. For study tools, we used a digital single-lens reflex (DSLR) camera with standardized settings, a dark box made of cardboard, and open-access RGB analysis software. Images were analyzed for differences in the numeric values of the RGB color and compared against the TSD. Correlations between TSD and age, gender, and environmental temperature were checked. This study involved 30 cases; these were analyzed and showed an increase in the numeric values of RGB for the corneal color as the TSD increases. Of note, the correlation of TSD with the color red was greater than for either blue or green; age had a positive correlation while gender had nearly no correlation, and the environmental temperature had a negative correlation. Based on this, gender was excluded from our equation. Also, we noted that the variance inflation factor of green was high and, therefore, excluded it from the predictive equation. The equation derived follows: TSD = {(0.091 x Age) + (0.171 x Red) + (0.018 x Blue) - (0.019 x Environmental Temperature) - 5.263}. Using this equation, the mean error was 21 minutes. This equation further narrowed the time range, usually given as four to six hours, when determining the TSD via conventional methods. Image analysis of corneal color after death using RGB analysis software can give us a more accurate and human error-free TSD that can be digitally stored and reproduced and, therefore, could prove useful in the forensic arena in the future.

Postmortem Ocular Findings in the Optical Coherence Tomography Era: A Proof of Concept Study Based on Six Forensic Cases

Postmortem analysis of the ocular globe is an important topic for forensic pathology and transplantology. Although crucial elements may be gathered from examining cadaveric eyes, the latter do not routinely undergo in-depth analysis. The paucity of quantitative and objective data that are obtainable using current, invasive necroscopic techniques is the main reason for the limited interest in this highly specialized procedure. The aim of the current study is to describe and to object for the first time the postmortem ocular changes by mean of portable optical coherence tomography for evaluating ocular tissues postmortem. The design involved the postmortem analysis (in situ, and without enucleation) of 12 eyes by portable spectral-domain Optical Coherence Tomography. The scans were performed, in corneal, retinal and angle modality at different intervals: <6 h, 6th, 12th, and 24th hour and after autopsy (25th-72nd hour). The morphological changes in the cornea, sclera, vitreous humor and aqueous humor were easy to explore and objectify in these tissues in first 72 h postmortem. On the other hand, the "in situ" observation of the retina was difficult due to the opacification of the lenses in the first 24 h after death.

Post-mortem ocular changes and time since death: Scoping review and future perspective

The forensics literature on the relationship between ocular changes and the postmortem interval (PMI) has traditionally focused on time-related variations in concentrations of metabolites and elements within the eye. However, structural changes within the eye over time deserve the same attention because there is significant evidence of their importance in determining the time since death. Therefore, we reviewed publications that provided substantial evidence on this issue. In light of our inclusion and exclusion criteria, 26 papers were eligible for review. On the basis of the ocular parameters considered, we grouped the reviewed evidence into eight thematic areas: corneal opacities, corneal thickness, drug-induced pupil diameter variations, retinal changes, segmentation of retinal vessels, intraocular pressure reduction, globe temperature and crystalline alterations. The most important and common limitations of the reviewed studies were small study populations (many were monocentric studies), absence of robust statistical methodology, use of mathematical models valid only in ideal conditions and validation only for short PMIs. Although many phenomena cannot be used to reliably estimate PMI, there is rigorous evidence suggesting that promising factors, including corneal thickness, require methodological innovations for application to forensics practice but could be used in the near future to reliably estimate the time since death.

Corneal-Smart Phone: A novel method to intelligently estimate postmortem interval

The changes of postmortem corneal opacity are often used to roughly estimate the postmortem interval (PMI) in forensic practice. The difficulty associated with this time estimate is the lack of objective means to rapidly quantify postmortem corneal changes in crime scenes. This study constructed a data analysis model of PMI estimation and implemented an intelligent analysis system for examining the sequential changes of postmortem corneal digital images, named Corneal-Smart Phone, which can be used to quickly estimate PMI. The smart phone was used in combination with an attachment device that provided a darkroom environment and a steady light source to capture postmortem corneal images. By segmenting the corneal pupil region images, six color features, Red (R), Green (G), Blue (B), Hue (H), Saturation (S), Brightness (V) and four texture features Contrast (CON), Correlation (COR), Angular Second Moment (ASM), and Homogeneity (HOM), were extracted and correlated with PMI model. The results indicated that CON had the highest correlation with PMI (R²  = 0.983). No intra/intersubject variation in CON values were observed (p > 0.05). With the increase in ambient temperature or the decrease in humidity, the CON values were increased. PMI prediction error was <3 h within 36 h postmortem and extended to about 6-8 h after 36 h postmortem. The correct classification rate of the blind test samples was 82%. Our study provides a method that combines postmortem corneal image acquisition and digital image analysis to enable users to quickly obtain PMI estimation.

Post-mortem Iris Decomposition and its Dynamics in Morgue Conditions

With increasing interest in employing iris biometrics as a forensic tool for identification by investigation authorities, there is a need for a thorough examination and understanding of postmortem decomposition processes that take place within the human eyeball, especially the iris. This can prove useful for fast and accurate matching of antemortem with postmortem data acquired at crime scenes or mass casualties, as well as for ensuring correct dispatching of bodies from the incident scene to a mortuary or funeral homes. Following these needs of forensic community, this paper offers an analysis of the coarse effects of eyeball decay done from a perspective of automatic iris recognition. We analyze postmortem iris images acquired for a subject with a very long postmortem observation time horizon (34 days), in both visible light and near-infrared light (860 nm), as the latter wavelength is used in commercial iris recognition systems. Conclusions and suggestions are provided that may aid forensic examiners in successfully utilizing iris patterns in postmortem identification of deceased subjects. Initial guidelines regarding the imaging process, types of illumination, and resolution are also given, together with expectations with respect to the iris features decomposition rates. Visible iris features possible for human, expert-based matching persists even up to 407 h postmortem, and near-infrared illumination is suggested for better mitigation of corneal opacity while imaging cadaver eyes (Post-mortem iris decomposition and its dynamics in morgue conditions. ArXiv pre-print, 2019).

[Temporal pattern of postmortem color changes in the pupil region of the cornea in rabbits]

OBJECTIVE

To explore the temporal pattern of postmortem color changes in the pupil region of the cornea for noninvasive estimation of the postmortem interval (PMI).

METHODS

Two rabbit models of air embolism and drowning were established in a dark room at a temperature of 20 ℃ with a relative humidity of 30%. The corneal images of the rabbits were acquired using a digital camera at two-hour intervals within 72 h after death. The pupil region on the corneal images was segmented using computer image processing technique (MATLAB), and the parameters of 6 image color features (RGBHSV) were extracted. Regression analysis was used to analyze the relationship between these parameters and the PMI, and the effects of different death causes on the changes of the corneal color features were also assessed.

RESULTS

Within 72 h after death from different causes, the R, G and B values of the pupil region on the corneal images all tended to increase with the PMI, showing a good fitting with the PMI (P &lt; 0.01). No significant correlation was found between the values of H, S and V and the PMI (P&gt;0.05). The R, G and B values in the pupil region on the corneal images showed consistent variation trends after death from the two causes, and their correlations with PMI were also similar. The measured values of R, G and B in air embolism group were greater than those in the drowning group.

CONCLUSIONS

The postmortem color changes of the pupil region on corneal images follow an identifiable temporal pattern and can vary across different causes of death. The regression equations established in this study provide references for non-invasive and objective estimation of the PMI.

A computational approach to estimate postmortem interval using opacity development of eye for human subjects

This paper presents an approach to postmortem interval (PMI) estimation, which is a very debated and complicated area of forensic science. Most of the reported methods to determine PMI in the literature are not practical because of the need for skilled persons and significant amounts of time, and give unsatisfactory results. Additionally, the error margin of PMI estimation increases proportionally with elapsed time after death. It is crucial to develop practical PMI estimation methods for forensic science. In this study, a computational system is developed to determine the PMI of human subjects by investigating postmortem opacity development of the eye. Relevant features from the eye images were extracted using image processing techniques to reflect gradual opacity development. The features were then investigated to predict the time after death using machine learning methods. The experimental results prove that the development of opacity can be utilized as a practical computational tool to determine PMI for human subjects.

Detection of granularity in dermoscopy images of malignant melanoma using color and texture features

Granularity, also called peppering and multiple blue-grey dots, is defined as an accumulation of tiny, blue-grey granules in dermoscopy images. Granularity is most closely associated with a diagnosis of malignant melanoma. This study analyzes areas of granularity with color and texture measures to discriminate granularity in melanoma from similar areas in non-melanoma skin lesions. The granular areas in dermoscopy images of 74 melanomas and 14 melanomas in situ were identified and manually selected. For 200 non-melanoma dermoscopy images, those areas which most closely resembled granularity in color and texture were similarly selected. Ten texture and twenty-two color measures were studied. The texture measures consisted of the average and range of energy, inertia, correlation, inverse difference, and entropy. The color measures consisted of absolute and relative RGB averages, absolute and relative RGB chromaticity averages, absolute and relative G/B averages, CIE X, Y, Z, X/Y, X/Z and Y/Z averages, R variance, and luminance. These measures were calculated for each granular area of the melanomas and the comparable areas in the non-melanoma images. Receiver operating characteristic (ROC) curve analysis showed that the best separation of melanoma images from non-melanoma images by granular area features was obtained with a combination of color and texture measures. Comparison of ROC results showed greater separation of melanoma from benign lesions using relative color than using absolute color. Statistical analysis showed that the four most significant measures of granularity in melanoma are two color measures and two texture measures averaged over the spots: relative blue, relative green, texture correlation, and texture energy range. The best feature set, utilizing texture and relative color measures, achieved an accuracy of 96.4% based on area under the receiver operating characteristic curve.